Skip to main content

Full text of "Maverick Ricardo Semler pdf"

See other formats

How would you like to work in a company that 
has no receptionists, secretaries, standard hier¬ 
archies, dress codes, or executive perks ... a 
company that lets you set your work hours and 
even your salary, and asks you to review your 
boss...a company whose ways of doing business 
are totally opposite those of most corporations? 
How would you like to work in a company that 
not only breaks all the rules, but succeeds? 

Can’t be done, you say? 

It can if you think like Ricardo Semler, 
the maverick CEO who turned his own com¬ 
pany into a model for the 90s and beyond. In 
MAVERICK, Semler will tell you how he 
changed his company and how you can change 
yours... if you dare! 

The extraordinary true story of 34-year- 
old CEO Ricardo Semler and of the rebirth of 
his family-owned Brazilian manufacturing 
company, Semco, MAVERICK is a chronicle of 
corporate change—a change so vast and suc¬ 
cessful that Semco has been visited by over 
400 corporate representatives from around the 
world who want to see it in action. 

In 1980 21-year-old Ricardo Semler took 
over the reins of the company his father founded 
and built. Ricardo proceeded to break every rule 
of doing business, firing an entire echelon of 
upper- level managers. Once he opened the 
floodgates to change, Semler never—even when 
he was tempted to—closed them again. 

Semler’s management system allows 
employees to work at home, study and discuss 

Digitized by the Internet Archive 
in 2017 with funding from 
Kahle/Austin Foundation 

The Success Story Behind the 
World’s Most Unusual Workplace 

Ricardo Semler 




A Time Warner Company 

The Cartoons in Ricardo’s A,B,C’s are reprinted 
with permission of Miguel Paiva. 

Copyright © 1993 by Tableturn, Inc. 
All rights reserved. 

Warner Books, Inc., 1271 Avenue of the Americas, New York, NY 10020 

| A Time Warner Company 

Printed in the United States of America 

First Printing: September 1993 
10 987654321 

Library of Congress Cataloging-in-Publication Data 
Semler, Ricardo. 

{Virando a propria mesa. English] 

Maverick : the success story behind the world’s most unusual 
workplace / Ricardo Semler. 
p. cm. 

ISBN 0-446-51696-1 

1. Semler, Ricardo. 2. Businessmen—Brazil—Biography. 
I. Title. 

HC186.5.S46A3 1993 





For Antonio Curt, whose soul resides at the head office, 
and for Sofia, who has taken my heart into perennial custody. 


First and foremost to Rick Levine, who devoted more wee hours to this 
book than he cares to remember. To Suzanne Gluck at ICM, who was naive 
enough to think we could hook a publisher; to Rick Horgan, Maureen 
Egen, and their colleagues at Warner Books; to Gail Rebuk at Random 
House U.K.; and to all the other publishers around the world for sharing 
our enthusiasm. Not to be forgotten are tough-outside, butter-inside 
Esther Newberg, Heather Shroeder, and the competent team of ladies at 
Colleridge's in London. And, of course, I am most indebted to the 
sixty-five people at Semco who let me interview them for so many hours, 
and to the hundreds of others at our company who made this story—and so 
many of our dreams—reality. 

Ricardo Semler 
Sao Paulo, 1993 

This is not a business book. It is a book about work, and how it can be 
changed for the better. It is based on the experiences of a company called 
bemco y which has managed to make money and improve the lives of the 
people who work for it. Some say Semco is merely a quirky laboratory run by 
a few impudent and iconoclastic managers. It isnt the real world, they say. 
Its Brazil. I ve heard all the other arguments by now,, too—we went too far, 

we went too fast , we are too big, we are too small, we are too high-tech, we 
aren t high-tech enough. 

But a few people—more excitable or impressionable, perhaps, but also 
more open-minded—will see in the story of this admittedly peculiar com¬ 
pany and its people a new way of running an organization. It is not 
socialist, as some of our critics contend. It isn’t purely capitalist, either. It is 

a new way. A third way. A more humane, trusting, productive, exhilarat- 
trig, and, in every sense, rewarding way. 




Every Wednesday afternoon dozens of men and women file through 
the front gate on their way to a third-floor meeting room at Semco, 
the company I lead in Sao Paulo, Brazil. The guard at the entrance 
has been expecting them. For years now, executives from some of the 
biggest and best-known companies in the world, IBM, General 
Motors, Ford, Kodak, Bayer, Nestle, Goodyear, Firestone, Pirelli, 
Alcoa, BASF, Chase Manhattan, Siemens, Dow Chemical, Mercedes- 
Benz, and Yashica among them, have been making an unlikely 

pilgrimage to our nondescript industrial complex on the outskirts of 
the city. 

Semco manufactures an impressively varied roster of products, 
including pumps that can empty an oil tanker in a night, dishwash¬ 
ers capable of scrubbing 4,100 plates an hour, cooling units for air 
conditioners that keep huge office towers comfortable during the 
most sweltering of heat waves, mixers that blend everything from 
rocket fuel to bubble gum, and entire biscuit factories, with 6,000 
separate components and sixteen miles of wiring. But it’s not what 
Semco makes that has executives and management experts the world 
over waiting months for a chance to tour our plants and offices. It’s 
the way the people of Semco make it. 

When I took over Semco from my father twelve years ago, it was a 
traditional company in every respect, with a pyramidal structure and 
a rule for every contingency. But today, our factory workers some- 

2 A Ricardo Semler 

times set their own production quotas and even come in on their own 
time to meet them, without prodding from management or overtime 
pay. They help redesign the products they make and formulate the 
marketing plans. Their bosses, for their part, can run our business 
units with extraordinary freedom, determining business strategy 
without interference from the brass. They even set their own salaries, 
with no strings. Then again, everyone will know what they are, since 
all financial information at Semco is openly discussed. Indeed, our 
workers have unlimited access to our books (and we only keep one 
set). To show we are serious about this, Semco, with the labor unions 
that represent our workers, developed a course to teach everyone, 
even messengers and cleaning people, to read balance sheets and 
cash flow statements. 

For truly big decisions, such as buying another company, everyone 
at Semco gets a vote. A few years ago, when we wanted to relocate a 
factory, we closed down for a day and everyone piled into buses to 
inspect three possible new sites. Then the workers decided. Their 
choice hardly thrilled us, since it was next to a company that was 
frequently on strike. But while no one in management wanted 
front-row seats to a labor-management war, we moved in anyway. 

In the lobby of our headquarters, a standard-issue office building 
with four floors of steel and glass, there is a reception desk but no 
receptionist. That’s the first clue that we are different. We don’t have 
receptionists. We don’t think they are necessary, despite all our 
visitors. We don’t have secretaries either, or personal assistants. We 
don’t believe in cluttering the payroll with ungratifying, dead-end 
jobs. Everyone at Semco, even top managers, fetches guests, stands 
over photocopiers, sends faxes, types letters, and dials the phone. We 
don’t have executive dining rooms, and parking is strictly first-come, 
first-served. It’s all part of running a “natural business.’’ At Semco 
we have stripped away the unnecessary perks and privileges that feed 
the ego but hurt the balance sheet and distract everyone from the 
crucial corporate tasks of making, selling, billing, and collecting. 

Our offices don’t even have the usual number of walls. Instead, a 
forest of plants separates the desks, computers, and drawing boards 
in our work areas. The mood is informal: some people wear suits and 

Maverick a 3 

ties or dresses, others jeans and sneakers. It doesn’t matter. If people 
want to emulate Thomas V^atson and don white button-downs, that’s 
fine. But turtlenecks and T-shirts are okay, too. And I want our 
people to feel free to put their feet on their desks, just like me. I am 
pleased to report that more than once a group of Semco executives 
has been interrupted by people who wanted to use their conference 
room to hold a birthday party. It warms my heart to see vice 
presidents eating cake on little plates decorated with Mickey and 

We have a sales manager named Rubin Agater who sits there 
reading the newspaper hour after hour, not even making a pretense of 
looking busy. I m sure this mystifies some of our visitors. Most 
modern managers wouldn’t tolerate it. But when a Semco pump on 
an oil tanker on the other side of the world fails and millions of 
gallons of oil are about to spill into the sea, Rubin springs into 
action. He knows everything there is to know about our pumps and 
how to fix them. That’s when he earns his salary. No one cares if he 
doesn’t look busy the rest of the time. 

My office is on the fourth floor—at least it was the last time I 
looked. I don’t use it as much as other proprietors. Most mornings I 
work at home. I concentrate better there, despite two sheepdogs that 
like to bark when I’m on the phone with important customers. I 
encourage other Semco managers to work at home, too. I also take at 
least two months off each year to travel, and I like to roam far. There 
are pictures in my office from two recent expeditions, a balloon safari 
in Tanzania and a trek through the Khyber Pass in Afghanistan. I 
never leave a number where I can be reached when I’m away and I 
don’t call in. I want everyone at Semco to be self-sufficient. The 
company is organized—well, maybe that’s not quite the right word 
for us—not to depend too much on any individual, especially me. I 
take it as a point of pride that twice on my return from long trips my 
office had been moved—and each time it got smaller. My role is that 
of a catalyst. I try to create an environment in which others make 
decisions. Success means not making them myself. 

One of my first acts at Semco was to throw out the rules. All 
companies have procedural bibles. Some look like the Encyclopaedia 

4 a Ricardo Semler 

Britannica. Who needs them? They discourage flexibility and com¬ 
fort the complacent. At Semco, we stay away from formulas and try 
to keep our minds open. I knew our rule book was useless when, as a 
test, I once distributed some additional pages for it. I asked some 
managers to read the new sections and give me their reaction. 
Almost everyone said they w r ere just fine. Trouble was, I had stapled 
the pages together so they couldn’t be read without first prying them 
apart. Funny how no one mentioned that. All that new employees at 
Semco get today is a twenty-page booklet we call The Survival 
Manual. It has lots of cartoons but few words. The basic message: use 
your common sense. 

If you haven’t guessed by now, Semco’s standard policy is no policy. 
Many companies have entire departments that generate mountains of 
paperwork trying to control their employees. Take travel. They have 
rules that govern how much a person can spend in every possible 
situation. At Semco, we want our people to spend whatever they 
think they should, as if they were taking a trip on their own, with 
their own money. There’s no department, no rules, no audits. If we re 
afraid to let people decide in which section of the plane to sit, or how 
many stars their hotel should have, we shouldn’t be sending them 
abroad to do business in our name, should we? 

We have absolute trust in our employees. In fact, we are partners 
with them. On the assumption that a capitalist society must be 
capitalist for all, Semco has a profit-sharing plan—but with a 
difference. Typically, companies hand down these plans like God 
handed Moses the Commandments. The owners decide who gets 
what, when. At Semco, profit sharing is democratic. We negotiated 
with our workers over the basic percentage to be distributed—about 
a quarter of our corporate profits, as it turned out—and they hold 
assemblies to decide how to split it. It’s up to them. Profit sharing 
has worked so well that once, during negotiations over a new labor 
contract, a union leader argued that too big a raise would overextend 
the company. 

Some people have likened the Semco philosophy to socialism, in 
the old, Eastern European sense. Nonsenseskaya. I think we’re proving 
that worker involvement doesn’t mean that bosses lose power. What 
we do strip away is the blind, irrational authoritarianism that 

Maverick a 5 

diminishes productivity. We’re thrilled our workers are self-governing 
and self-managing. It means they care about their jobs and about 
their company, and that’s good for all of us. 

In restructuring Semco, we’ve picked the best from many systems. 
From capitalism we take the ideals of personal freedom, individual¬ 
ism, and competition. From the theory, not the practice, of socialism 
we have learned to control greed and share information and power. 
The Japanese have taught us the value of flexibility, although we 
shrink from their familylike ties to the company and their automatic 
veneration of elders. We want people to advance because of compe¬ 
tence, not longevity or conformity. 

When you eliminate rigid thought and hierarchical structure, 
things usually get messy, which is how our factories look. Instead of 
machines neatly aligned in long straight rows, the way Henry Ford 
wanted it, they are set at odd angles and in unexpected places. That’s 
because our workers typically work in clusters or teams, assembling a 
complete product, not just an isolated component. That gives them 
more control and responsibility, which makes them happier and our 
products better. Nearly all our workers have mastered several jobs. 
They even drive forklifts to keep teammates supplied with raw 
materials and spare parts, which they have been known to purchase 
themselves from suppliers. 

The Metalworkers’ Union at first resisted this flexibility. Long 
ago, organized labor was forced to adopt narrower and narrower 
job classifications as a defense against giant corporations that 
pushed ever harder for higher productivity and profits. Eventually 
the unions realized that they could turn the system against the 
corporate masters by refusing to allow any deviation from the 
rules without extra pay. With time, the system became more bene¬ 
ficial to labor than to management—but it really wasn’t serving 
either side. 

When the union realized that Semco had no intention of dismantling 
its power, that the higher profits our factories would generate would 
mean higher pay for its members, and that we were intent on giving 
workers a meaningful say in our business, obstructionism eased. We 

6 a Ricardo Semler 

have been allowed to innovate—to let our employees innovate. We 
are all freer. 

Our factory workers, for example, can come in anytime between 7 
a.m. and 9 a.m. It’s their choice, not ours. What if one worker 
wants to start at 7:00 and a teammate decides to come in at 9:00? 
Surely that would disrupt production? That was our worry, too, so 
we set up a task force to mediate any problems. It hasn’t met yet. 
Our workers knew that production would suffer if they didn’t 
coordinate their schedules, so that’s what they did. 

At Semco we don’t even like to think in terms like worker or boss. 
We prefer Associate and Coordinator. And we encourage everyone to 
mix with everyone else, regardless of job. In our offices the purchas¬ 
ing and engineering departments have been scrambled so that 
everyone sits together, near the factory. The idea is that we all can 
learn from one another. The office decor is anything but plush and 
sometimes there is grime and oil on the floor. No one cared but me, 
and now that I have stopped wearing my fancy Italian loafers with 
the paper-thin soles, I don’t care either. 

What do the bosses think of all this? I am often asked. 

Well, we don’t have as many bosses as we used to. As workers 
began to exercise more control over their jobs and assume more voice 
in our policies, the need for supervisors diminished. We have also 
reduced our corporate staff, which provides legal, accounting, and 
marketing expertise to our manufacturing units, by more than 75 
percent. We no longer even have data processing or training depart¬ 
ments. Everyone vouches for his own work, so we don’t need a 
quality control department either. After taking a good look at 
ourselves, we whittled the bureaucracy from twelve layers of manage¬ 
ment to three and devised a new structure based on concentric circles 
to replace the traditional, and confining, corporate pyramid. 

We’ve also changed the way our departments do business with one 
another. If one doesn’t want to buy services from another, it’s free to 
go outside the company and buy from someone else. The threat of 
competition keeps us all on our toes. Recently, we have encouraged 
employees to start their own companies, leasing them Semco machin¬ 
ery at favorable rates. We buy from our former employees, of course, 

Maverick a 7 

but they are also free to sell to others, even Semco’s competitors. 
This program has made us leaner and more agile, and given them 

ultimate control of their work lives. It makes entrepreneurs out of 

This is an extreme case, of course, but we try to maximize the 
possibilities and minimize supervision for everyone at Semco. Not 
that we don’t have accountability. Before people are hired or pro¬ 
moted to leadership positions, they are interviewed and approved by 
all who will be working for them. And every six months managers 
are evaluated by those who work under them. The results are posted 
for all to see. Does this mean workers can fire their bosses? I guess it 
does, since anyone who consistently gets bad grades usually leaves 
Semco, one way or another. 

We aren't the only company to experiment with participative 
management. It s become a fad. But so many efforts at workplace 
democracy are just so much hot air. Not that the intentions are bad, 
it s just that it s much easier to talk about worker involvement than 
implement it. We have been ripping apart Semco and putting it back 
together for a dozen years, and were just 30 percent finished. Still, 
the rewards have already been substantial. 

We’ve taken a company that was moribund and made it thrive, 
chiefly by refusing to squander our greatest resource, our people. 
Semco has grown sixfold despite withering recessions, staggering 
inflation, and chaotic national economic policy. Productivity has 
increased nearly sevenfold. Profits have risen fivefold. And we have 
had periods of up to fourteen months in which not one worker has 
left us. We have a backlog of more than 2,000 job applications, 
hundreds from people who say they would take any job just to be at 
Semco. As a matter of fact, our last help-wanted newspaper ad 
generated more the 1,400 responses in the first week. And in a poll 
of recent college graduates conducted by a leading Brazilian maga¬ 
zine, 25 percent of the men and 13 percent of the women said Semco 
was the company at which they most wanted to work. 

Not long ago the wife of one of our workers came to see a member of 
our human resources staff. She was puzzled about her husband’s 
behavior. He no longer yelled at the kids, she said, and asked 

8 A Ricardo Semler 

everyone what they wanted to do on the weekends. He wasn t his 
usual, grumpy, autocratic self. 

The woman was worried. What, she wondered, were we doing to 
her husband? 

We realized that as Semco changed for the better, he had, 



The home movie shifts abruptly to a scene of a narrow dirt road on 
the outskirts of Sao Paulo. It is the late 1950s. The camera moves in, 
shaking perilously as home movies will, and three men and a truck 
come into view. In the truck bed lies a horizontal lathe, a humble 
tool used lor cutting metal. From the reverential expression on the 
face of the lean, suntanned man on the left, you would think it was 
fashioned from gold. 

It is a glimpse of what it means to start a business. Few moments 
were as gratifying to my father, that gaunt man in the movie, as 
when the lirst machine was delivered to his new plant. 

Antonio Curt Semler was born in Vienna in 1912, at the ignoble 
end of a glorious empire. His father, a successful dentist, wanted 
Antonio to be a physician. But he had his heart set on engineering, 
and was graduated from Vienna’s Polytechnic University instead. 

Sensing an imminent decline of opportunities in between-the-wars 
Austria, he looked for challenges abroad. In 1937, Curt answered a 
want ad placed by the DuPont Company and soon accepted a position 
as production engineer at its chemical and textile plant in Argentina, 
where he survived the political upheavals of Peronism and eventually 
became plant manager. 

In 1952, Curt visited Brazil and was intrigued by the prospects 
such a vast, undeveloped country presented. At the time, he was 
working on a patent for a centrifuge that could separate lubricating 

10 a Ricardo Semler 

oil from vegetables and, he hoped, would be the mainstay of his own 
business. At first Rio de Janeiro seemed the right spot. But he 
changed his mind upon viewing Sao Paulo from a hill overlooking 
the city. This was a place, he thought to himself, for an entrepreneur. 

It was also the place where this forty-two-year-old bachelor would 
meet another Austrian exile, Renee Weinmann, marry and raise a 
family. My mother’s father had moved his family to China and 
started a profitable business exporting textiles and porcelain from 
Shanghai. The Weinmanns lived a sheltered life, alongside thousands 
of Europeans for whom the British and French quarters of Shanghai 
were islands in an ocean of misery. But as the poverty of the Chinese 
deepened, their resentment of foreigners grew. My grandfather died 
just before the Communists took over and my grandmother and her 
only child left on a freighter, practically expelled by Mao Ze-dong. 
They landed at the port of Santos, in the south of Brazil, after a 
month at sea. Soon, my mother was giving English lessons to 
wealthy citizens of Sao Paulo, one of whom introduced her to my 

On the dining room table of their small two-bedroom apartment 
in the center of town, my father plotted strategy for Semco, a 
contraction of Semler & Company. He worked constantly, pausing 
only to allow my mother to reclaim the table so they could dine 
together. Then he would work late into the night. 

Children were a priority, for Curt was already anxious for someone 
to take over the business. Three pregnancies ended in failure before 
my sister, Susan, was born. But it was hard for my father to imagine 
a woman running a rough-and-tumble business such as Semco. I was 
born only eleven months later, an unknowing heir to an enterprise 
scarcely older than I was. 

Semco and its owner were obscure, as was the oil-separating 
machine for which my father had by this time secured a patent. But 
Brazil was a land of opportunity, ravenous for all sorts of new 
products. In no time the Semco centrifuge became a market leader, 
and three Brazilian partners had come aboard with badly needed 
capital and a knowledge of the country. My father kept working 
hard, and the 1950s and 1960s responded with more kindness. These 
were the years of the Brazilian Miracle, when the economy was 

Maverick a 11 

spurting by 7 percent annually, a rate unmatched anywhere. Semco, a 
backyard machine shop only a few years earlier, blossomed into an 
established manufacturer employing about 110 people and generat¬ 
ing revenues of $2 million a year. 

The military dictators who ran Brazil, although strictly capitalists, 
must have been impressed with Stalin: they loved five-year plans. In 
the late 1960s, they decided the country needed a shipbuilding 
industry and suddenly there was a five-year National Shipbuilding 
Plan. My father, sensing opportunity, took on two British pump 
manufacturers as partners and Semco became a major supplier of 

marine pumps in Brazil, a line of products for which it is still 

Passing the company to the next generation is the stated goal of 
entrepreneurs everywhere, although it’s sort of mystifying when you 
think about it. Have you heard of Einstein’s son? Would you expect 
to see the evening news presented one day by Dan Rather’s son? Why 
is it that the offspring of industrialists are typically expected to be 
natural entrepreneurs? Sometimes a son or daughter of an entrepre¬ 
neur can be an excellent manager, as Thomas Watson, Jr., of IBM 
proved in the 1960s and 1970s. But who makes the succession 

decision? Invariably, it is the father. And let’s be honest: he isn’t 
exactly impartial. 

”1 want the business to remain in the family’s bosom,” the tycoons 
say. (Remind you of Louis XIV?) So-and-so is “the only one I can 
trust. But is it better to have the company sink slowly in trustwor¬ 
thy hands than prosper through the efforts of strangers? 

In my case, I hardly inspired much confidence. As a child, I had 
an attention span that made jai alai seem like a chess match. I 
thought it was perfectly normal to do math homework with cartoons 
dancing across the television screen. My mother still can t understand 
how they let me graduate with so many Ds, or why she and my 
father were so frequently summoned to meetings with teachers to 
discuss, among other things, my belief that peanut butter and jelly 
sandwiches were made to hang from cafeteria ceilings. 

As I grew older, I came to feel downright ashamed of my 
unwavering need to occupy stage center. But some powerful current 

12 a Ricardo Semler 

kept pulling me toward any situation that held the potential for 
attention and leadership. In high school, I pursued and attained 
many positions—class president, captain of the track team, photog¬ 
raphy editor of the yearbook. I even ran the lunchtime snack stand, a 
concession operated by the junior class to raise money for the class 
trip the following year. It was here that an aptitude for business first 
surfaced. I abandoned the policy of freebies honored by past snack 
stand impresarios, lengthened the hours of operation, and encouraged 
suppliers such as Coke, Pepsi, Kellogg, and Frito-Lay to compete for 
our business. I wouldn’t say that they fell over one another for the 
snack stand account, but our revenues soared. Then I invested the 
profits in the stock market, and by the following year had earned 
enough to fly the entire class to a resort. “At school on business 
only,” was how my classmates described me in the yearbook. 

But this nascent capitalist flair paled beside my desire to master 
the guitar, in particular a Gibson Les Paul. I was obsessed with rock 
n roll—the Stones, Led Zeppelin, Pink Floyd, Uriah Heep. To me, 
diversification meant also learning to play the bass and the drums. 
All this alarmed my father, who anxiously awaited the end of what he 
hoped was a phase to begin talk of my corporate ascension. 

When I was sixteen I took a summer job in Semco’s purchasing 
department and found myself studying specs for the half-inch steel 
plates on marine pumps. My only fun came when a trucker taught 
me to drive. I was the only kid on my block to get to slide behind 
the wheel of a Mercedes, even if it was a ten-ton truck, not a 560 
SEL convertible. 

After the summer I went back to the guitar and my despairing 
father considered selling out. Soon enough, though, I discovered the 
depressingly narrow range of my musical abilities. Given my father’s 
desire for a successor and my stint as a snack stand mogul, business 
seemed a sensible aspiration. 

There was just one problem: In Brazil, all healthy young men 
must serve in the army for at least a year when they turn eighteen. 

Having waited this long, my father thought it would be the best 
medicine for me. I would be compelled to get into shape and, if not 
learn to love the dawn, at least develop a passing acquaintance with 

Maverick a 13 

it. Whether he was just giving me a hard time or meant it I never 
discovered. Just in case, I began lobbying my mother. Mealtimes 
became sacred, grades improved, and my bed was miraculously made 
with regularity. Weekends were inaugurated by early-morning rising, 

with the help of two alarm clocks, both positioned on the far side of 
my room. 

Fathers understand this sort of behavior, and mine kept giving me 

a you-can’t-trick-me-son” look. Mothers are another story, God bless 

Curt, mine would say. Dickie just can’t lose a year of his life in 
the army.” 

I would nod vigorously in the background. 

It 11 make him a man, my father would reply. “I was in the 
army myself, you know.” 

My mother and I would sit back on the sofa to listen (again) to 

tales of marches in the snow, cold showers, and the polishing of boots 
and belt buckles. 

Soon I was ordered for a physical at the local recruiting office. Off I 
went in the middle of the night to sit in line with all the other guys. 
Any important physical or mental defects were considered then, and I 
watched several people in the line practice flat feet, nervous break¬ 
downs, and assorted malfunctions. I didn’t have the courage for that. 
The recruiters liked my height, were impressed with my weight, 
and, generally, didn’t find me lacking in any of their stringent 
qualifications, such as having two legs and two arms. 

Two weeks later it was time to present my army-approved phy¬ 
sique at another military office for a more complete medical exam. 
Friends of mine had told me horror stories about it. According to 
tradition, all who were dismissed from service were invited (by a 
soldier whose physique would have made Sylvester Stallone jealous) 
to donate blood for the army hospital. An attendant would stick a 
huge needle in your forearm, hand you a two-liter bottle, and move 
on to the next victim. Having passed out on 100 percent of the 
occasions on which I have had any contact with blood, I could 
already picture it: the slow darkening before my eyes, my body 
lurching forward, the sound of breaking glass, a vision of sandy 

14 A Ricardo Semler 

beaches in the South Pacific. Then a heavy head and many soldiers 
frowning at me from the ceiling. 

Well, if that was the price of freedom . . . 

Off I went before dawn, and after the usual five-hour wait found 
myself in a line with hundreds of naked young men, answering 
questions about childhood diseases, adulthood diseases, and, espe¬ 
cially, venereal diseases. One guy put on an impeccable show of 
tongue swallowing and was immediately placed under arrest. Once 
again I passed. At least I didn’t have to give blood, I told myself. 

A month later I went for the written examination. By now getting 
up at 4 a.m. was easy, since I wasn’t getting any sleep anyway. The 
test was long and complicated. I did my best to do my worst and 
went home in a state of high anxiety. Three days later I learned I had 
passed again. 

1 was already outfitted in uniform, cap, and boots when I was 
summoned by the officer in charge and, after waiting two and a half 
hours and then explaining why I thought it was silly to be in an army in 
a country that has never fought a war and hasn’t the slightest chance 
of breaking that record, was dismissed. 

Passing up business school, I entered the Sao Paulo State Law 
School, the most prestigious university in the country and one that 
had educated many Brazilian presidents. I was sure 1 would prefer its 
more humanistic approach. But I barely scraped by with below - 

average grades, in part because I started working again at Semco soon 
after I enrolled. 



My first experiences in the executive suite distressed me. Everyone 
was as starched as their shirts. 

I tried to fit in, I really did. I even went to a trendy mens store 
and acquired a complete corporate outfit—navy blue suit with white 
pinstripes, white shirt with French cuffs, black shoes. I didn’t wear 
the suit—the suit wore me. 

I hadn’t been at work long when my father summoned me to his 
office and told me I shouldn’t let anyone call me by my nickname, 
Dickie. It would diminish respect for me, he said sternly. I should 
always be known as Dr. Ricardo, a title that college graduates such as 
me are automatically accorded in Brazil. Never mind that I had 
already asked everyone to call me Dickie. The signals to the 
organization became so confused that to this day I still get memos 
addressed to Dr. Dickie. 

Like most self-made men, my father was a traditionalist. He 
treated his employees paternalistically and considered strikes and 
labor strife personal affronts. He always remembered that he had 
founded Semco on a dining room table. 

He would wake up every day precisely at 6:30 and follow a 
schedule as rigid as an I-beam. On Tuesdays, Wednesdays, and 
Thursdays he would play an early-morning round of golf before going 
to the plant, lunch at home, then nap for fifteen minutes before 
returning to the office. He would arrive for dinner precisely at 7:45 

16 A Ricardo Semler 

p.m. On Fridays he would skip golf and attend the Rotary Club 

His secretary, Fernande, a short, stocky woman with a serious 
demeanor, had already been with him fifteen years when I arrived, 
though I never understood why, since they were always complaining 
about each other. If Fernande arrived six minutes late, my father 
would not fail to comment. If he corrected a mistake on a letter with 
a pen instead of a pencil, it would be her turn to fume. 

Fernande would often charge out of my father’s office banging the 
door loudly behind her, and I could understand why. An elegant man 
with an imperial bearing—he had been a member of the Austrian 
Olympic ski team and still carried himself like an athlete, with his 
chest pushed forward and his shoulders straight—he demanded 
respect and not infrequently inspired fear. A stern look was his 
registered trademark. When I was growing up friends would often 
leave me at the front door rather than risk an encounter with the old 
man. At the office, clerks would flip coins to decide who would 
deliver his papers and possibly become the target of his wrath, for he 
would often take out his frustration on whoever happened to be 
closest to him. 

When I got my own office next to his, he insisted on installing a 
sliding door between us. I would be in the middle of a meeting with 
a customer and—wham—the door would open and in he would 

My father didn’t like my habit of putting my feet up on my desk. 
He also didn’t like my habit of working at home. And it annoyed 
him that I took pains to separate my personal life from the company. 
If I made a copy of something for my own use, I’d always pay for it. 
If I needed to send a telex to a hotel where I planned to spend my 
vacation, I’d pay for that, too. To my father, the business and the 
family were intertwined and inseparable. 

The gap between us was almost fifty years wide, and no matter 
how hard we tried to blend them, our styles and our ideas would 
become increasingly incompatible. I knew the transition from father 
to son wasn’t going to be smooth. Slowly, he came to realize it as 

Maverick a 17 

The tensions between my father and me were compounded by an 
economy that had shifted into reverse. Despite the generals’ five-year 
plan, Brazil s shipbuilding industry was among the hardest hit 
sectors. This was devastating for Semco, for by 1980 90 percent of 
our business was in such marine products as pumps, components for 
propellers, and water-oil separators for ship motors. The only profits 

Semco was generating were from investments of cash reserves, which 
were diminishing rapidly. 

Semco had been run for more than a decade by a group of 
executives well connected to the maritime industry. As the slump 
deepened, 1 became convinced Semco’s only chance for survival was 
to broaden its product line and reduce its dependence on the marine 
business. It was time, as the Monty Python crew used to say, for 
something completely different. But whenever I brought up diversifi¬ 
cation, they would argue that Semco was a highly focused company 
and that if we lost our specialized skills, we would lose everything. 
The old guard kept telling my father that Semco could hang on, that 

a new shipbuilding plan was about to start and we would bounce 

Semco was about to take on $ 1 million in financing to build a new 
plant with computer-controlled machinery to assemble marine pump 
casings and propellers. I was once enthusiastic about such an ad¬ 
vanced factory, but as the economy slid I worried about making such 
a huge investment. If we were going to take on more debt, at least 
let it be to enter a healthy industry. But I was only'twenty and, 
despite my last name and the inflated title on my business card, 
assistant to the board of directors, I didn’t have much influence. 
Whenever a critical decision approached, I would usually be sent on 
a meaningless visit to some out-of-town customer, keeping me away 
from the office until they figured out what to do. 

I was convinced I wanted to be in business—the Les Paul was 
consigned to the bottom of the closet by now—but I wasn’t sure 
Semco would be the place at which I would realize my entrepreneurial 
aspirations. Pushed by me, Semco’s auditors at Price Waterhouse had 
found some companies they thought we might be interested in 
acquiring. Few seemed synergistic with Semco, but there was one 
that caught my eye. It couldnt have been a more mundane business—it 

18 a Ricardo Semler 

made ladders, all sorts of ladders. Started by a carpenter, it had 
quickly become the largest ladder maker in Brazil, with more 
employees than Semco. But it had fallen on hard times, so much so 
that Price Waterhouse refused to be responsible for the numbers in 
its pre-acquisition audit. 

Undeterred, I began to think about leaving Semco and going into 
the ladder business on my own. I started visiting the ladder maker 
nearly every day, getting to know the company and whittling away at 
the owner’s asking price, which started out at more than $1 million, 
despite the company’s problems. 

As the price dropped, I tried to convince my father to lend me the 
money. “Let’s think about it,” he would say. “Show me the balance 

I did, but he wouldn’t look at it. He was stalling. 

It had nothing to do with ladders, of course. He didn’t want me to 

I spent most of a year studying the ladder company, and it 
proved a much better business education than I got in any class¬ 
room. I talked to the firm’s creditors. I talked to its suppliers. 
I milked Price Waterhouse for all the strategic thinking I could 

Then I had a brainstorm. I took out a newspaper ad, in English, 
announcing that I was looking for a CEO for my prospective 
enterprise. I said I wanted someone who had managed a Brazilian 
subsidiary of a multinational company. 

I was swamped with responses, perhaps because I hinted that I was 
willing to give up some equity in the company. I met with about 
thirty of the most impressive candidates. I would go through the 
numbers and then ask them to tell me how they would revive the 
company. By the time I was finished with the interviews I had not 
only found a CEO—a Bolivian who was president of Black & Decker 
Brazil—but had also become quite wise. 

Ladder sales had continued to, well, slide, and the company’s price 
at last reached a level I could afford: $1. The company’s suppliers 
were willing to extend me additional credit, besides. Though young, I 
was connected to a well-known company. Perhaps they thought Semco 

Maverick a 19 

was behind the deal, an assumption I somehow never got around to 

We drew up an exceedingly complicated contract, which was my 
lawyers' way of warning me that it was a high-risk move. But I 
reasoned that if I were going to go broke, I would rather it be when I 
was in my twenties than in my forties, when I might have additional 
responsibilities in short pants. Anyway, thanks to all those interviews, 
I had a plan. I was going to discontinue nearly half the company’s 
products, take all the fresh credit the suppliers would give me and 

buy enough time to rebuild the business around the items that were 
still profitable. 

On the day we were finally ready to close the deal my father 
summoned me to his office. 

“Let’s talk about this some more,” he said. 

Its too late, I told him. ‘Tve been at this for a year.” 

I knew what was troubling him and tried to ease his anxiety. I told 
him I would join Semco’s board of directors and come to the office 
whenever he wanted. “But there’s no use me staying here,” I added. 
“We’re just going to fight. We don’t agree on how to run the 
company, and you ve got all these people here you still trust. There 
was a time when I thought you would give me more power, but I 
know you’re not about to do that.” 

That s when my father finally said he wanted to discuss a change 
in ownership that would give me a majority of Semco’s shares and 
with them the authority to make the changes I believed were needed 
if the company was going to endure. He also agreed to pay the 
$30,000 the contract stipulated I would have to pay if I walked away 
from the ladder company deal. (That company, incidentally, lingered 
for a while, no doubt because the owner eliminated some products I 
would have abolished. But in the end the company was liquidated. 

As for my future CEO, he became the director of DuPont’s Brazilian 

“Who’s going to run Semco?” I asked my father, just to hear him 
say it. 

I was quiet. He was quiet. 

Better make your mistakes, he finally said, “while I’m still 

20 A Ricardo Semler 

As it happened, I made all sorts of mistakes right under his nose. 
That’s the trouble with mistakes. You don’t recognize them at the 
time. And there was one move that wasn’t a mistake, although it was 
terribly risky. Either I could continue to try to convince the 
entrenched executives that Semco had to change, and endure more of 
their foot dragging, or 1 could make the changes without them. The 
executives still believed in the shipbuilding plan and were prepared 
to wait for it, collecting their comfortable salaries all the while. So it 
would be the second route. 

I m leaving on a trip for the next two or three weeks,’’ my father 
said toward the end of our difficult conversation. “Whatever changes 
you want to make in the organization, make them now.” 

I settled into the office my father had used and reviewed the yellow 
legal pad before me. On it I had scrawled the names of all our top 
executives. There were about fifteen names in all. It was a Friday, 
and I had set up individual appointments with each of them, starting 
with Waldemar Maragonni, an imposing man with piercing eyes. He 
greeted me icily, letting his eyes bore through my young skull. I felt 
like a student handing in a term paper that was three days late. 

Hello, Waldemar,” I stammered. 

“Dickie, what can I do for you?” Waldemar said condescendingly. 

Come on, Dickie. Say it. Say it. 

Well, you see, Waldemar, 1 think we’ve got to make some 
changes around here.” 

No reaction. 

“You know that I’ve been wanting to diversify Semco.” 

Still no reaction. 

“And,” I went on, “I feel that we have to make some management 
changes for this diversification process to work.” 

Silence. I decided to summarize the situation thus far. Surely he 
would agree with me in the end, and we could part cordially, if not 
as chums. So off I went into medieval history, tracing Semco’s 
beginnings and Waldemar’s contributions in his decade and a half at 
the company. Alter a while it occurred to me that I was making a 
good case for a promotion, not a dismissal. Abruptly changing tack, 

Maverick a 21 

I rushed into the past few years, concentrating on Semco’s gloomy 
outlook and the need for new products. 

“I couldn’t agree more," Waldemar finally said. Ah, progress. But 
no, he’s not supposed to agree with me. 

“We have started to study diversification,” Waldemar went on. “I 

myself will be in Germany and England next month to talk to a few 

But Waldemar. You ve said that many times, and nothing has 
happened so far.” 

“Now it will be different.” 

No arguing with that. An hour and a half had gone by. I had only 
three hours left in the afternoon, and a long list of managers to fire. 

Waldemar, let s make a long story short. I want to conduct this 
diversification myself, and I can’t see that happening with a dual 

There. I ’ve done it. 

“No problem,” Waldemar replied, to my chagrin. “We’ll make a 
great team.” 

I felt the stirring of panic, followed by tension and then frustra¬ 
tion. Were all the meetings going to be like this? What if the 
afternoon ended before I got through the list? The survivors would 
surely organize over the weekend. 

“Waldemar, let me be as straightforward as possible. I need to do 
this on my own.” 

Silence descended like a thick wool blanket on a hot summer day. 

There we sat, air conditioner whirring in the background, looking at 
each other. 

“Are you saying I’m fired?" Waldemar said after a while. 

I cleared my throat and composed myself. “Yes, Waldemar, I am. 
It’s the only way.” 

More silence. Then the inevitable question. 

“Does your father know about this?” 

“Well, yes. Sort of. Yes. He does. Yes.” 

I see, Waldemar said. At last I saw a physical reaction: he 
became pale, then ghostly white. I wondered how I looked. Probably 

22 a Ricardo Semler 

Waldemar got up. “Well, I’ll be in my office Monday morning. 
Next week we can discuss the transition.” 

“Ah, Waldemar, one minute please,” I quickly interjected, getting 
up also. “You see, I have in mind a fast transition.” 

“And what would that be? Six months?” 

“Uh, no, no,” I stammered. 

“Three months, then?” Waldemar said, moving toward the door. 

“Well, no, less. . . I. . . I. . . I . . .” 

“You don’t mean you want me out in less than a month?” His eyes 
narrowed as they stared into mine. 

“Well, actually, yes.” 

“When do you intend that I leave the company?” 


“Waldemar, I’d like you to take your things home today. Should I 
need something on Monday, I’ll call you there.” 

Waldemar went through the doorway, glanced back, then turned 
and shut the door, firmly. I collapsed on the sofa. 

The rest of the meetings were over in a hurry. By 6 p.m. I had 
fired 60 percent of Semco’s top management. I’d never hired any¬ 
body, much less laid waste to so many people in a single, Godfather¬ 
like purge. But they had ignored my suggestions to diversify. Semco 
did not have time for slow, herbal medicine. It needed emergency 

With a mixture of relief and anxiety I made my way to my car and 
drove through the gates on my way home for the weekend. 

I chose a Friday because I had hoped everything would simmer 
down over the weekend. I had much organizing to do before Monday. 

I had no managerial experience, but had already lined up someone 
who had. My new right arm went by the name of Ernesto Gabriele. 
An excessively thin man in his forties, Ernesto had enough energy for 
an entire executive committee (and would practically be one). Born 
to an Italian immigrant family in Sorocaba, a small town in the 
interior of the state of Sao Paulo, he had held fast during his youth to 
the dream of studying in America. Back then, he didn’t speak a word 
of English or have the money for an airline ticket, much less tuition. 
But he was determined, and that proved sufficient. 

Maverick a 23 

Near his hometown was the Brazilian subsidiary of the Firestone 
Tire company, and at a school in Sorocaba Ernesto found applications 
for the University of Akron, where the rubber giant was based. High 
school diploma in hand, Ernesto filled out a form, requesting a 
scholarship and promising to learn English. The school had an 
opening for a foreign student and Ernesto was admitted. He collected 
donations from the townspeople for his ticket, studied English long 
enough to confirm his suspicion that he could pick up languages 
almost overnight, and when July rolled around took off for Ohio. 

Four years later he had a degree from the University of Akron and 
a job next door as a Firestone trainee. Rising quickly, he was soon on 
his way back to Brazil as a junior manager. He kept on rising, too, 
becoming at the age of twenty-nine Firestone s youngest vice president. 

Ernesto specialized in administration, finance, and restlessness. 
Highly ethical but short-tempered, he was quick to take offense or 
create showdowns with colleagues from which he would extricate 
himself by handing in his resignation. He became vice president of 
J. I. Case, which makes tractors, then moved to a subsidiary of a 
German multinational, where he learned to speak German. His next 
stint was at a French company, where he learned French. Then he 
became the assistant to the president of Sharp, the Japanese electron¬ 
ics company, but left after screaming bullshit in Japanese during a 
business meeting. He went to work in Rio as a vice president for 
administration for Xerox, but after making eleven trips to the United 
States in one year decided he yearned for a less hectic life. Leafing 
through the newspaper in his Rio apartment one Sunday, an ad, in 
English, caught his attention. It was for a position as CEO at a 
ladder company in Sao Paulo. 

Ernesto’s resume intrigued me. How could it not, since he had 
listed all the companies he had worked for, including ones at which 
he had spent only a couple of months, along with a complete list of 
references from each? His openness, coupled with his former employ¬ 
ers’ willingness to recommend him, impressed me. He had shot up 
early in his career but had not found what he was looking for. He was 
obviously brilliant, with a passion for organization, but he also had a 
strong character and inevitably became unhappy wherever he went. 
That fit my needs not for the ladder company but for Semco. We 

24 A Ricardo Semler 

were headed for uncharted waters, and I anticipated we would have 
to change course often. A traditional executive might balk at that, or 
prove too inflexible. 

I met with Ernesto several times before the Great Purge and hired 
him the Saturday after it. Now I had an alter ego who was as 
hardheaded and crazy as I was—but he knew how to run a company. 

When Monday arrived I began to realize the impact of what I had 
done. Clients would call and ask for a particular person and we 
would explain he was no longer with us. They would ask for someone 
else, and we’d tell them he wasn’t here, either. After their fourth or 
fifth try, you could hear their bewilderment. 

Our former executives, I realized, had made themselves the 
repository of many of our trade secrets, no doubt to insure their 
continued employment. They didn’t count on anyone as rash as me 
spoiling it for them, but I didn’t count on people as secretive as they 
were. We had to search through their desk drawers and file cabinets 
to discover all the special deals and arrangements they had made with 

Naturally, some of these customers tried to talk us into taking the 
executives back. A German company that licensed us to make several 
kinds of marine pumps even sent two representatives all the way to 
Brazil to personally plead with me to step aside and let the old 
managers return. They made it clear they thought we wouldn’t 
survive without them. 

I wasn t sure we would, either. But we were going to try. 



We spent the rest of 1980 gasping for oxygen as we ran from bank 
to bank, crying to raise cash. We even considered selling the 
company’s buildings. It was no fun having to prepay bills from 
suppliers or sell our products in a market full of rumors about 
Semco. But soon Ernesto began to work his multinational magic and 

our customers realized we weren’t going to go bankrupt, at least not 
right away. 

Ernesto installed dozens of new procedures and invented new 
forms almost daily. Our resident managerial wizard convinced sales¬ 
people to fill out customer-visit reports and keep statistics on orders 
closed versus quotes offered. Files were rigorously organized all over 
the company. Employees’ bags and cars were searched under a system 
that randomly selected them for inspection. Everyone was issued a 
plastic ID card and compelled to wear it. Production schedules were 
displayed on boards in our new planning and control department. 
Members of our new time and methods department were dispatched 
around the plant, searching for ways to speed workers up. 

How inspiring. How capitalist. Everything Xerox could do, Semco 
could do, too. Okay, so they actually sold their products and we only 
had the statistical evidence of why ours were still in the warehouse. 
A mere detail, we thought, to be rectified any day. 

We promoted many Semco old-timers, some of whom are with us 
still. But we sorely lacked an aggressive leader for our sales depart- 

26 A Ricardo Semler 

ment. Truth was, we lacked almost everything in our sales depart¬ 
ment. Brazil’s five-year plans, with their long wind-ups, had made 
us lax. There seemed no need to sell anything; we would take orders, 
that’s all. Which is why in 1980 we were delivering pumps that we 
sold in 1975. 

So I put an ad in the newspaper for a sales manager. (Sound 
familiar?) And that’s how 1 met Harro Heyde. 

Harro was a strange duck. He was very tall, balding, and wore 
spectacles with lenses like the bottoms of Coke bottles. He didn’t 
own a suit and his clothes seemed pulled at random from the bottom 
of some dark, dank closet. At our first interview he told me he was 
an avid cook (coq au vin was his specialty) and that he and his wife 
(a parachute instructor, incidentally) lived with two small children in 
a Bavarian-style cottage in the exurbs, baking bread and raising 
(what else?) chickens. 

Harro had been first in his class at the Institute of Aeronautical 
Technology, the MIT of Brazil. He spoke English, French, and 
German, and had held an assortment of jobs, including maintenance 
director at a huge mining complex in the Amazon and construction 
manager for the French company Fichet in Algeria. His credentials 
seemed sufficiently zany for what I had in mind, which was to shake 
up our sales department. 

Harro was willing. Like Ernesto, he had yet to find his niche. 

Harro started selling on his first day at Semco and never stopped. 
He would pass a salesperson’s desk and ask how a given quote was 
progressing. The salesperson would tell him the customer was 
studying it. Harro would suggest a visit. The salesperson would say 
that he would try to arrange one for the next week. Harro would say 
he didn’t mean the next week, he meant now. Impossible, the 
salesman would argue. But Harro would grab him by the arm, and a 
bunch of magazines from our reception area, and head straight for 
the prospect’s office. Announcing their arrival, they would wait one, 
two, even three hours—whatever it took until they were seen. 
Meanwhile, Harro would keep himself entertained by talking to the 
receptionist about the company, grilling the Semco salesperson about 
his other accounts, or reading the magazines he brought with him. 

I did what I could to keep him supplied with reading matter. A 

Maverick a 27 

few days after Harro joined Semco, we signed a license agreement 
with a Norwegian company to manufacture a new line of pumps. We 
knew very little about the technical specs of these pumps, but in a 
few days we were scheduled to make a major presentation on the 
pumps at Petrobras, the state-owned oil company that is among the 
world s largest. I dropped by Harro’s office and asked him if he 
would look at the material. No problem, he said. Before he could 
change his mind, I handed him 600 fun-filled pages on hydraulics. 
En passant, I told him I wanted him to speak to the Petrobras 
engineers. Harro just gave me a leave-it-to-me-kid smile. 

He spent dozens of hours studying the stuff, spoke often to the 

Norwegians, then made a pitch to Petrobras and came back with a 
substantial order. 

Harro and 1 hit the road. In two years, we traveled to sixteen 
countries and contacted more than sixty companies to drum up 
business. We would visit a dozen, sometimes two dozen cities in a 
month. To protect us from burnout, our rule was one flight per day. 
We broke it often. On one unforgettably neurotic trip, we gulped 
down a dinner of marinated salmon in Oslo in an open-air restaurant 
where the midnight sun reflected off our silverware; took the next 
flight to New York, where we lunched with executives from the 
Crane Company, then on to dinner in Cincinnati with representatives 
from the Day Mixing Company; before spending the night (in the 
dark, this time) in San Francisco, where we met the next morning 
with executives from Pacific Pumps, Inc. 

I came to appreciate Harro s rowdy jokes and respect his appetite. 
Once we headed for Linz, Austria, to visit a pump company called 
Ochsner. Adr. Ochsner himself was to receive us, but something went 
wrong and we didn’t get to the hotel until close to midnight, six 
hours late. Harro immediately made his way to the restaurant, on a 
quest for his favorite Austrian fare, knoedel, a kneaded ball of dough 
that is a potato substitute. The waiter was apologetic, but we were to 
leave Austria the next morning and Harro was not an easy man to 
dissuade. Off he went to the kitchen, not only securing a promise 
from the chef that his wish would be fulfilled but also that he would 
be allowed to watch the preparations. 

28 A Ricardo Semler 

During these trips we hoped we could convince enough companies 
to let us manufacture their pumps and mixers under license in Brazil, 
augmenting our pitiful cash flow and buying us time. But the 
negotiations were difficult; Brazil's quirky economy and restrictive 
government regulations were bad enough, and Semco’s relative 
obscurity made our task that much harder. We slipped and slid as we 
danced along the edge, but someone up there was watching out for 
us, for we managed to secure seven license agreements to manufac¬ 
ture soup mixers, oil filters, liquid agitators for mineral tanks, 
compressors, and other industrial and food-processing equipment. 
All this would reduce marine equipment to 60 percent of our 
product line. We were still breathing. 

Early in 1981 Alcoa announced plans to build a new aluminum 
mill in the north of Brazil. It was one of the few large construction 
projects we could expect in the midst of the recession. Semco bid on 
the 200 hydraulic pumps for the project, a $3 million job. An order 
like that meant more hope for our company, which normally had a 
backlog of ten months of production work but now was down to a 
mere two and a half months. We had supplied Alcoa before and 
thought our chances of doing so again were good. But late one Friday 
we were told that an Australian competitor, Warman Pumps, had 
underbid us. 

Recovering from the disappointment (as much as possible), we 
turned to our last hope: an order for marine pumps from two local 
shipyards, Caneco and Emaq, which were building three ships for 
Petrobras. Both shipyards told us they would rather buy from Semco, 
as was their custom. Trouble was, one of our competitors had been 
instructed by its foreign parent to win the job at any cost, and had 
offered an 18 percent rebate. They were desperate, too. 

Backs to the wall, we offered a 20 percent discount. There 
wouldn’t be much profit in the job, but the cash would help keep us 
going until we could figure out what to do next. I remember feeling 
quite pleased on the flight back from Rio. 

At the end of the week, the shipyards told us the competition had 
offered a 35 percent discount and thrown in inspections, insurance, 
and shipping costs free. Do you know the medical expression 

Maverick a 29 

flatline ? Well, it fit us. Had there been an electrocardiogram 
machine in our board room, it wouldn’t have registered a heartbeat. 

We were on the next flight to Rio, of course, to match the offer. 
We had no choice. This was surely the last order of the year, and it 
was only March. Price was secondary—we had to survive until we 
could pull Semco together. We matched the 35 percent discount and 

the shipyards gave us their word the order was ours. Our hearts were 
pounding again. 

A few days later we were sitting at a meeting. I don’t remember 
what we were discussing, but I do recall how jittery we all were. One 
of us, a superstitious soul, noticed there were thirteen people around 
the table. At that precise moment I was summoned for an urgent 
telephone call. 1 picked up the receiver and was told that the Caneco 
shipyard had been offered an additional 7 percent discount by our 
rival and had accepted. Emaq was going along, since Petrobras 
wanted all six ships to be identical. 

All of us were crushed. Except Harro. He had that gleam in his 
eye. At his urging, we had decided to open another front in our 
battle for survival. We would win the Alcoa job after all. 

Alcoa had already reached deals with nearly all the suppliers it 
needed to build its plant. The only bid it had not yet decided was for 
the giant mixers that rotate slowly to keep the aluminum from 
hardening at the bottom of the tanks. We knew little about these 
machines, except that they were huge, with blades fifteen feet long 
and shafts three stories high. We asked Alcoa if it would consider a 
bid from Semco. They told us we might as well forget it, since the 
winner was to be chosen in just ten days. 

The next afternoon Harro and 1 took off for King of Prussia, 
Pennsylvania, home of one of the largest mixer manufacturers in the 
world, the Philadelphia Gear Corporation. \^7e needed its expertise to 
pull off the job, and it would need Semco, since under our trade laws 
the machines would have to be manufactured in Brazil. We were 
ushered into the president s office, where he told us his position on 
Brazil, sparing no unflattering adjectives or adverbs. Philadelphia 
Gear once tried to buy a Brazilian company but had been frustrated 
by the horrendous red tape. From then on, he wanted nothing to do 
with Brazil or Brazilians. 

30 A Ricardo Semler 

When I mentioned Alcoa his scowl disappeared. Philadelphia Gear 
did business with the biggest aluminum manufacturers, but not 
Alcoa. He immediately saw the chance to establish a relationship 
with the aluminum giant through a back door in South America. He 
was smiling now, but turned pale again when we told him the mixers 
required were bigger than any his company had built. When we told 
him that we had only a few days to finish the technical work, 
negotiate the deal, and sign the license agreement, his scowl returned. 
Even so, after consulting with his board he decided to go ahead. 

Back in Brazil, Harro camped out in the reception area of Alcoa’s 
imposing offices. The procurement manager, Victor Barruzzi, was a 
tough man, with no interest in cordialities. He was also an early 
riser, so Harro made it a habit to arrive at his office at around 6 a.m. 
Barruzzi would pass by and barely nod. Harro would announce 
himself every hour or so and just sit there, reading Alcoa annual 
reports and mining magazines. Barruzzi finally gave in and saw him, 
but only to say that Semco didn’t have a chance for the job. Harro 
somehow managed to set up another meeting with Barruzzi for both 
of us and, after waiting two hours, we were received. We begged and 
badgered until Barruzzi gave in and let us bid. 

And that s how Alcoa s new Brazilian plant came to boast twenty- 
six Semco-Philadelphia mixers, including four that are among the 
largest in the world. 

Now that we had a winning streak, we decided to attack the 
Petrobras problem head on. One man, Shigeaki Ueki, the president, 
was responsible for the oil company’s decisions. After countless 
requests we managed to secure a meeting with him on a Friday 

afternoon. We told him we believed our competitor was dumping_ 

selling pumps at a loss to worm its way into the Brazilian market. 
Prove it, he said, and the order was ours. But, he warned, we didn’t 
have much time. 

The needle on that electrocardiogram machine jumped. Then we 
remembered that it was almost quitting time and we still had to get 
a copy of our competitor’s balance sheet from the Official Registrar. 
The clerk who received us clearly did not wish to spend the last half 
hour of her work week looking through musty files, but somehow 

Maverick a 31 

my eighteen-days-in-the-desert expression moved her. She retrieved 
the balance sheet and I scanned it with apprehensive eyes. What if a 
creative accountant had camouflaged the losses? He might be playing 
around with inventories, or future earnings. How could I tell? My 
fingers swept down the columns. Earnings, earnings, where were 
those damn earnings? Ah, there they were. Nice, fat numbers, in 
parentheses. As I suspected, a $3 million loss. 

In the end, Dr. Ueki played Solomon and divided the pump order 
in half. Semco would have the contract for three of the ships. 

If only to show how the world goes around and around but ends 
up where it started, this order, which we believed was essential for 
our survival, was delayed several times. Then the shipyard went 
broke before it finished paying us. 

But that was much later. For the moment, the noose was a little 

Nineteen eighty-one came and went. And 1982. Thanks to the 
license agreements Harro and I secured, Semco found itself in the 
unaccustomed position of making money. Lots of money. Well, what 
for us was lots of money. We even had some cash left over after we 
paid off our debts. With Ernesto’s new corporate controls in place, 
and our product line greatly enlarged, we were riding a wave of 
success as Brazil continued to founder. 

I couldn t have been more pleased with the new, improved, 
diversified Semco. Everything seemed so professional. No one could 
get in or out of our plant without showing an ID card. Even I was 
stopped by our guards. We had special forms for overtime, phone 
invoices, copying machine receipts, everything. 

Our pride and joy was our new budget system. The numbers were 
ready on the fifth working day of every month, all in color-coded 
folders. How much coffee was consumed by workers in a subsection 
of Light Manufacturing III? There it is on chart No. 112, page 67. 

Ernesto was making us over in the image of Firestone, Xerox, and 

When I talked to executives at other companies about our system, 

I liked to tell them about the time the president of Allis-Chalmers 
came down from the United States for a visit. He wondered whether 

32 A Ricardo Semler 

we might want to acquire his company’s pump operations in Brazil. 
The deal didn’t work out, but we took him on a tour of our factory 
anyway. He leafed through some of our budget folders and seemed as 
impressed as a child at his first circus. He didn’t expect to find such 
efficiency—compulsiveness?—at a Latin American company. He told 
us he was going to order his Brazilian subsidiary, which was many 
times larger than Semco, to install a similar system right away. Our 
chests stayed puffed for weeks. 

Not too long after that, Allis-Chalmers started losing market share 
and then money. It has now been dissolved and sold piecemeal. 

We would come to see a lesson in that. But meanwhile, we were 
anxious to share our new management skills with the world. It was 
time for Semco to go out and buy some companies. 



I began to suspect we weren’t being taken seriously when Ray 
Krinker, senior partner of Price Waterhouse in Brazil, looked across 
the table at me and smiled. “A grocery store near my home,” he said. 

That s the best buy in your price range.” 

It was 1983. In our typically brash way we had called Krinker, a 
specialist in mergers and acquisitions, and told him we had $500,000 

to spend. We wanted to buy a company. A grocery store wasn’t what 
I had in mind. 

I remembered a certain ladder maker I could have bought for $1 
and implored Krinker and his people to press on. They did, and 
during the next few months I looked at stacks of documents from 
divisions and subsidiaries that were on the block. We had several 
requirements: the company had to be No. 1 or No. 2 in its market, 
have a connection with our own businesses, and be technologically 
advanced. It also had to be for sale for the right reasons, which were 
either because there was no one to succeed the founder, current 
management was inefficient, or the parent company had lost interest. 
We weren’t in the market for companies that made products of 
dubious quality or required large infusions of cash, which, of course, 
we didn’t have. 

Krinker, an outspoken man with eyes that were alternately twin¬ 
kling and intense, was invaluable in our quest. Also ego-leveling. “A 
small hole,” he was fond of telling me, ‘‘can sink a big ship.” 

34 A Ricardo Semler 

I can t say we were a big ship, but we sure as hell were looking at 
many small holes. The companies that were up for sale then— 
remember, this was still during a Brazilian recession—weren’t very 
healthy. Most lacked either capital or a market. Some lacked both. 
Ray helped us avoid them. But that still left plenty of candidates. 
We would sit for whole afternoons working our way through foot- 
high piles of annual reports. We would pull one out, read about the 
product line, look at the numbers, and then decide whether it 

belonged in the small pile of interesting propositions or much larger 
stack of rejects. 

No company with half a million dollars has ever felt more 
powerful. Here were firms with annual sales of $5 million, $10 
million, even $20 million, awaiting our verdict. We even got 

literature from one company with $150 million in sales and another 
with 4,500 employees. 

Of course, we never considered talking to outfits that big. Even 
so, we had no idea how pretentious we were in thinking we had the 
expertise to run other companies besides our own. We were a legend 
in our own minds. Who would dare stop us from fulfilling our 
destiny as a multinational conglomerate? 

I would think back on these times years later, when I read about 
the whiz kids of Wall Street playing at the same game, buying, 
selling, and merging companies, oblivious to the rich, complex 
histories each organization had. These Armani-outfitted, BMW- 
driving know-it-alls broke corporations into pieces in the name of 
short-term profits, and broke the hearts of those whose dreams the 
corporations embodied. But there was no place for sentiment or 
tradition or employee motivation in their world. “Bottom line, 
what’s the bottom line?” they bellowed into their cellular phones. I 
had become caught up in that way of thinking, foolishly believing I 
held the power of life and death over the companies whose vital 
statistics were gathered in the folders on our meeting table. Ray 
Krinker was right. We should have started slowly. We should have 
proceeded cautiously. But we were impatient, so we didn’t. 

We selected fifteen companies to visit, negotiated with six, and 
settled on one: a subsidiary of a Swedish air-conditioning company 
now known as the Asea Brown Boven Group. Its Brazilian offshoot, 

Maverick a 35 

Flakt, sold refrigeration equipment for ships and offshore oil-drilling 
platforms and ventilation systems for marine engine rooms. 

Negotiations proceeded quickly. The deal entailed taking over the 
Brazilian marine operation and making it a division of Semco. The 
price was about $300,000, which was in our ballpark. (Okay, it was 
almost our entire ballpark.) The signing ceremony was to be in 

Off I went, stopping to visit other Flakt plants on the way. Given 
my state, which was anxious and jet-lagged, I suppose it wasn’t 
really surprising that the trip was also the occasion of my first 
encounter with socialized medicine. For months I had been waking 
up with a throat so sore it was impossible for me to swallow even 
liquids before midmorning. I also suffered blinding headaches and 
gastritis. An extra fifty pounds completed the dismal picture. I 
hadn’t exercised for more than five years, unless you count huffing 
my way up the three flights of stairs to my office. 

I arrived in Goteborg the evening before an 8 a.m. meeting. My 
host called to say my hotel was a six-minute walk from the office, so 
naturally he would meet me in the lobby at six minutes to eight. I 
had a light dinner in my room and lined up the pills (two blue, three 
red-and-white, and one all-white) my doctors had prescribed. I took 
them, painfully because of my throat, set an alarm, and arranged for 
a wake-up call at 7:00, just in case. Then I went to sleep. 

In the midst of a dream I heard the distant ring of a telephone. I 
answered it in my dream, but it kept on ringing. Moments later I 
realized, to my despair, that it was six minutes to eight and the 
phone beside my bed was ringing. I fumbled for the receiver. 

Good morning, Mr. Semler,” I heard a crisp voice say. “I’m in the 

Yeah . . . um . . . right, ’ I stammered. An excuse. I needed an excuse. 
Thinking I was being clever, I explained that I would be delayed 
because I was awaiting a most important business call from Brazil. I 
went off to splash some cold, Swedish water on my face, only later 
realizing that it was 3 a.m. in South America. 

That day I visited an Asea Brown Boveri factory. As we walked 
through the plant, I felt progressively faint. Then, suddenly, every¬ 
thing went black. I had passed out. I was taken to the medical office 

36 a Ricardo Semler 

and after an hour or so felt well enough to resume my tour. I was 
feeling fine the next night, when I attended a dinner at the 

company s headquarters in Stockholm and signed and celebrated our 

We took over Flakfs three-floor factory in the suburbs of Rio de 
Janeiro and half its sixty-person payroll immediately. But our under¬ 
standing of the business turned out to be more limited than we had 
thought and we proceeded to lose more than $1 million during the 
next four years before the division turned a profit. 

At the time, though, we thought we were hot. So, naively pleased 
with ourselves, we continued our hunt for companies to buy. 

We sniffed around another outfit in the same field as Flakt, 
Baltimore Aircoil (BAC), a subsidiary of Merck Sharp and Dohme, 
the huge pharmaceutical house, but were told we were too late. A 
deal had almost been completed to sell the company to the largest 
air-conditioning contractor in Brazil. 

We figured the Merck subsidiary was worth $2 million to $3 
million. We only had $200,000. But talk is cheap, so we sent a telex 
to Merck suggesting a meeting. 

Merck told us it had already been offered close to what it wanted 
and asked if we were planning to bid more. Since we didn’t have 
much cash, we offered Merck an advance payment of sorts—the cash 
in their own subsidiary's till—and proposed to pay off the rest of the 
purchase price over five years, under a formula that called for fixed 
annual payments plus 25 percent of the unit’s profits. A Merck 
executive—a tall, wiry guy with the cold, hard eyes of a chief 
financial officer—let out a cynical laugh. But I noticed that another 
Merck man was looking on thoughtfully. I even thought I saw a 
slight smile form at the edges of his mouth. As it turned out, some 
of the Merck people had already concluded they might be better off 
with the brash young upstarts at Semco. 

Extended negotiations followed, ending in a complex contract that 
was signed at Merck’s lawyers’ offices in downtown Sao Paulo. As we 
waited for the final copies of the documents, we heard loud noises in 
the street below. From the window, we saw a huge crowd demonstrat¬ 
ing against the government’s economic policy. Suddenly, police riot 

Maverick a 37 

squads arrived and began firing tear gas canisters. Protesters were 
running all over as the caustic fog spread. One American looked at 
another. “What’s taking those papers so long?’’ he said. 

The next day we took charge of the BAC plant in the nearby town 
of Diadema. It was a large, gleaming factory, with flags flying above 

the entrance. Of the ninety people who worked there, we would 
retain more than sixty. 

Since the company was doing well, we left the old management in 
place, adding only a financial comptroller. The man we chose for the 
job, Antonio Carlos Iotti, was a rotund five-footer with thick 
eyeglasses that, he told me, he even wore in the shower. At his 
previous job with a large sewing machine company lotti had been 
ordered by a boss to redo an inventory count. He refused. When he 
received an angry memo from the boss demanding the recount, he 
sent an even nastier note back. The boss immediately went to see 
lotti, intent on at least a shouting match. He stood at least a foot 
taller than his insolent employee, but when he lowered his eyes he 
saw the diminutive Iotti, hands on his hips, glaring up at him, ready 
for anything. Iotti was the man for us, I thought. 



In two deals we had spent our $500,000 nest egg, nearly doubled 
our work force, and tripled the number of plants. But with our new 
financial controls we believed our managers could handle anything. 
Semco was doing so well I should have known we were heading for a 
bump in the road. 

Make that three bumps. The first had to do with Ernesto. 

Years earlier my father had invited two foreign companies to 
become shareholders in Semco, each with 24.5 percent of the stock. 
(He kept the rest, and control.) Semco had the right to manufacture 
their products, and the two companies had representatives on our 
board of directors. 

My father had a fine relationship with one, but no end of trouble 
with the other, which accused him—unfairly—of charging family 
expenses to the business and paying himself lavishly to avoid distrib¬ 
uting more profits in dividends. 

As Ernesto professionalized the company, he made every effort to 
be absolutely impartial. That meant giving Semco’s minority 
shareholders—including the troublesome one—as much information 
as he gave my family. After our second acquisition, Ernesto decided 
to send out a letter giving his opinion on our overall financial 
situation. He went into great detail on our investments and produc¬ 
tion plans. None of the information was particularly sensitive, but he 

40 A Ricardo Semler 

hadn t discussed it with my father, who felt so betrayed he stopped 
talking to Ernesto. 

I sided with my father, though I came to see that my familial 
loyalty betrayed my professional immaturity. I guess Semco was still 
very much a family business. 

I called Ernesto in and told him my feelings. He replied, in his 
hotheaded way, that he saw no solution other than his immediate 
departure. He turned and left for his office, picked up his things, 
and hurried down the stairs. I stood by the window and watched him 
walk through the plant gate, realizing that he wasn’t even going to 
take his company car home. I rushed to the parking lot, got into my 
car, and followed him. When I caught up with him, I saw he was in 
tears. I opened the passenger door and he got in. Both of us were 
silent as I drove him home. It was one of the saddest moments of my 
life, but it was merely a prelude to the real tragedy, for shortly 
afterward Ernesto was killed in a motorcycle accident. 

Ernestos replacement was a short, thin young man named Fernando 
Lotamorro. Another self-made man, Fernando had started as a 
traveling salesman and by his late twenties had become a top manager of 
a large Brazilian company, where he earned a reputation as a first-in, 
last-out guy. A proponent of meticulous financial controls, he was 
the ideal person to carry on and expand Ernesto’s good work. Or 
what we then believed was good work. 

The second bump involved Harro. He liked the bustle of Sao 
Paulo, but he loved the country. He would only live in a place where 
he could raise chickens, which meant his home was an hour-and-a- 
half commute through world-class traffic to the plant. 

As word of his exploits spread, Harro began getting job offers. He 
didn’t seriously consider them until he got a call from a French- 
Brazilian electrical equipment company, Merlin Gerin, which wanted 
him to run its plant in the rural state of Santa Catarina. He could 
live and work in a small town where a traffic jam meant beeping the 
cows out of the way and he would have room for all the chickens he 
could ever want and 5,000 roses in his garden. 

Harro asked me what I thought. I had to agree the offer was too 
good to pass up. He stayed at Semco just a few more months before 

Maverick a 41 

moving south. (Postscript: We kept in touch and ten years later Harro 
would rejoin us in a fashion, taking a 25 percent stake in a new parts 
and maintenance company we would start together and returning to 
work at our headquarters, where the unit is based.) 

No one could replace Harro, but sometimes things have a way of 
balancing out. About the time he left, we welcomed a newcomer 
who would quickly become a major force behind the astonishing 
changes that were about to take place at Semco. 

I knew our new, tough-minded, statistical approach, along with 
our acquisitions and all the new employees, had created a lot of 
organizational stress. Ordinarily it would have been silly for a 
company our size to have a human resources director, but I was 
convinced we needed one. So I hired a headhunter, went through 
dozens of resumes, and interviewed the ten most impressive candi¬ 
dates. All came from much bigger companies; none seemed to 

Enter, through a side door, Clovis Bojikian. Born to Armenian 
immigrants, Clovis was an idealist who started out as a schoolteacher. 
A proponent of Summerhill-style progressive education programs, he 
became a counselor at a school that served as a training ground for 
newly graduated teachers from the University of Sao Paulo and, 
eventually, its director. His tenure was marked by constant innova¬ 
tion. students were taught to think, not merely to memorize; they 
were encouraged to question everything, free to set their own 
schedules, dress as they pleased, and choose their own curriculum. 

But Brazil was still ruled by the generals, and they considered 
Clovis’s work subversive. Besides his suspicious lack of discipline, he 
was teaching kids to be skeptical of history texts that chronicled the 
exploits of those same military rulers. In 1984, the generals assem¬ 
bled a special tribunal at the university and Clovis testified for 
forty-six hours. Then he was fired. 

The students immediately occupied the school, demanding Clovis’s 
reinstatement. Armed with brooms and in some cases Molotov 
cocktails, and sustained by cans of food, they spent more than a week 
on the barricades and the front pages. The generals, embarrassed, 
finally had the army invade the school. The protesters, rounded up at 

42 A Ricardo Semler 

midnight by machine-gun-toting soldiers, marched out singing the 
national anthem. 

Clovis moved on to the Ford Motor Company in Sao Paulo, where 
he worked in personnel and training. After eighteen years he had 
risen to become Ford’s human resources manager in Brazil, but in the 
end such a megacorporation didn’t leave enough room for innovation. 
Clovis moved on to be human resources director at KSB Pumps, 
which makes hydraulic water pumps and is one of Semco’s competi¬ 
tors. Again, he wasn’t given much space to stretch. 

Here s where the story gets a bit cosmic. One of the members of 
my father’s old guard had moved to KSB and took an immediate 
dislike to Clovis. This former Semco executive hired the headhunting 
firm that got him his job to find a new position somewhere else for 
Clovis. That firm turned out to be the same one I had retained to 
Find Semco a human resources director. 

Clovis and I met in my office. Headhunters prefer their candidates 

to wear blue suits, white shirts, neat ties, and be clean-shaven. 

Yet here was an excruciatingly thin man in his late forties wearing 

an ugly brown suit, a beige shirt, and a white mustache that 

wound upward at the edges, a la Buffalo Bill. (At least he had 
a tie.) 

I liked him as soon as I saw him. 

Clovis would later reveal that he didn’t pay much attention to me 
at first. After all, he was supposed to have an interview with the 
president of Semco, and I was the age of his youngest child. Even 
after I sat down and began a conversation, he figured I was just 
keeping him company until the real president showed up. 

We spoke for almost three hours, and almost from the start we 
were completing each other’s sentences. We agreed then and there 
that Clovis would begin as our new human resources director the 

next Monday. It took about a week for both of us to realize that we 
hadn’t discussed his salary. 

The third bump? About a year after Clovis joined us he was offered 
a job as vice president of one of Brazil’s largest financial companies 
where he would run a department with hundreds of employees and 
earn almost three times as much as he did with us. As with Harro 

Maverick a 43 

we talked about the offer. It was a tempting one, and I told Clovis 

so. I also said that, should the job disappoint him, we would 
welcome him back. 

I spent almost a year interviewing replacements, without success. 
Then I stopped looking and simply waited. A short while later 
Clovis was back at Semco, earning his old salary. 

He gets offers all the time these days, which isn’t surprising, given 
what we have accomplished. Thank God he isn’t budging. 



By the middle of 1984 we were—you guessed it—ready to buy 
another company. By this time we had more or less recovered from 
our first, scary years and our diversification was working. We had no 
bank loans, still had some cash in the till (enough for working 

capital, anyway), and were already expanding slightly. It was a solid 
position, financially. 

We heard that Booz, Allen & Hamilton, the international consult¬ 
ing firm, had been hired to sell off Hobart Brazil, a subsidiary of the 
American conglomerate Dart & Kraft that manufactured dishwash¬ 
ers, meat slicers, and other food-processing equipment. With an 
80,000 square foot plant and 150 employees in Brazil, it was way too 
big for us. We asked for the acquisition documents anyway and were 
told we should move quickly. Is that all they ever say? 

Looking through the material, we came across Dart & Kraft’s 
annual report. To this day, I believe we were the only potential 
bidders to read it. Mixed in with pictures of the conglomerate’s 
products, including Duracell batteries, Tupperware containers, and 
umpteen varieties of cheese, were hints that told us Hobart’s success 
came from the marketing synergy of its dishwashers, fryers, scales, 
slicers, and related products. It was easier to sell the whole line than 
to try to move the items individually; the company would get more 
shelf space at distributors, lor starters. Other suitors were offering to 
buy only part of the whole; some wanted the dishwasher line, some 

46 a Ricardo Semler 

the scale business. Only Semco offered to buy everything. Only 
Semco wanted to maintain the synergy. 

The sum of the other bids reached Dart & Kraft’s asking price. 
Our proposal amounted to not much more than half that figure, and 
because of our lack of cash we would be paying it off over six years. 
But it turned out that Dart & Kraft was more interested in 
maintaining the Hobart image, since that name would be on the 
appliances it would continue to sell elsewhere. Keeping the Brazilian 
company intact, rather than breaking it up product by product, was 

We found ourselves around a long table at the Brazilian office of 
Price Waterhouse, completing the deal. We had a list of the issues to 
resolve, including the price of the specific assets, royalty percentages, 
and interest rates. \X^e told them the only item we were willing to 
increase was the time between our payments. (Ha, ha, ha.) 

The chief Dart & Kraft executive listened to us attentively. When 
we finished our presentation, there was silence. Everyone turned to 
him. He meditated awhile, then uttered these words: “I may be 

plump and have rosy cheeks, but I’m not dressed in red and it’s not 

After the laughter subsided, we bargained for seven hours, reduc¬ 
ing the payout period and increasing the royalties we would pay. 
Then we shook hands on a deal. 

Hobart Brazil’s Ipiranga plant, near Sao Paulo, had a long but less 
than glorious legacy, endured by a weary band of employees, some of 
whom had invested as many as forty-five years there. There was 
much too much?—history there, which made the changes we 
would eventually observe all the more surprising. 

Built in the 1930s, it was one of those gloomy gothic factories 
with high ceilings and tiny windows that would have been at home 
in Sheffield, England, or Pittsburgh, Pennsylvania. I don’t know who 
built it, but by 1939 it was owned by the Hobart brothers, 
proprietors of a successful restaurant equipment business based in 
Ohio. In those days the mechanical scales the plant cranked out were 
the most popular in Brazil, despite thirty local competitors. But over 
the years so many managers were sent down from Ohio, and so few 

Maverick a 47 

had any knowledge of the local market, that success slipped away. 
Hobart executives had a reputation for keeping to themselves. They 
weren’t interested in wining and dining distributors all over our vast 
country. And they were tethered to Ohio, dependent on the home 
office for even minor decisions, slowing their reactions and making 
them less flexible than they should have been. 

The chief beneficiary of these shortcomings was Filizola, Hobart’s 
main competitor, which though founded by Italian immigrants still 
managed to build strong ties with distributors across Brazil. Filizola 
expanded steadily and by the 1970s was impinging on Hobart’s 
balance sheet. The folks in Ohio had two options: invest more in the 
Brazilian division, buying new equipment that would eventually 
reduce costs and prices, or fight it out with whatever firepower was 
already there and hope for the best. Hobart chose the cheaper 
strategy and Filizola’s market share kept rising. 

By the end of the decade a local Hobart executive, Keith Rae, was 
promoted to general manager. Rae had been with the company more 
than twenty years, many of them as a salesperson in Brazil’s wildest ter¬ 
ritories. A man with an unfailing sense of humor and an unflagging 
honesty, Rae managed to keep the company afloat, but it’s a wonder 
how. He made constant trips to Ohio to report to his supervisors and 
spent another two months each year drafting and submitting a 
budget. All this cost him the time he needed to improve the 
products and attract new customers. 

What finally ended Hobart s Brazilian adventure was the elec¬ 
tronic scale. By the end of the 1970s it was difficult to sell a 
mechanical scale in the United States. Still, Hobart’s home office 
thought there would be a market for the more primitive models in 
the developing world and saw no urgency in retooling the Brazilian 
plant. Eventually Rae convinced management that the electronic 
scale was inevitable, even in South America. Hobart Brazil proudly 
announced its new product was on the way, but then failed to deliver 
it on time. Meanwhile, Filizola introduced its own electronic scale, 
establishing technological superiority and cornering the market. 

About this time the Hobart family decided to sell out to Dart & 
Kraft. This meant the Brazilian subsidiary could no longer rely on 
family members who had a long history in South America. Now the 

48 A Ricardo Semler 

operation would be evaluated on cold, impersonal parameters. Dart 
& Kraft had cash to invest, but would do so only if the products and 
the market showed promise. That did not seem the case with Hobart 
Brazil, which had by then slid to fifth place in its field. 

Dart & Kraft decided that of the nine offices and plants Hobart 
Brazil operated all but two should be closed. The plant at Ipiranga 
was spared, but the classic cycle of fewer sales, diminished invest¬ 
ment, and even fewer sales was under way. 

Besides scales, the plant produced most of the other Hobart 
products, including meat grinders and slicers, which sold decently, if 
unspectacularly, and dishwashers that were still a leader in the hotel, 
restaurant, and cafeteria market. But after a few years, when the 
leaner Hobart Brazil proved no more successful than its predecessor, 
Dart & Kraft hung out a For Sale sign. Keith Rae and his people 
were both dismayed and relieved. A new owner might not want to 
keep them; then again, a new owner might make the investment that 
they so badly needed and dissipate the shroud of failure that now 
hung over the place. 

The Monday after we closed the deal we were all at our new food 
service equipment plant—Fernando, Clovis, and I. Gloom had given 

way to anxiety. Everyone wanted to know what the new bosses were 

Our first move was to reaffirm the philosophy born with our first 
acquisitions ol not changing anything we didn’t understand and 
giving the people already in place a chance before supplanting them 
with outsiders. It s so easy to blame the managers when a business 
does badly, but often they haven’t had the freedom to manage or the 
motivation to perform as if the business were theirs. 

Keith Rae, by then in his sixties, was promoted to vice president. 
(To our great sadness, he died about a year after we bought Hobart.) 
We also had high hopes for Joao Vendramin, a Hobart engineer who 
had been trained as an economist. Shy and thoughtful, he would 
make a superb plant manager, we believed. Another old-timer, Joao 
Martins, had been engineering manager, plant manager, and market¬ 
ing manager in his forty-two years at Hobart Brazil, but the bosses 
in Ohio thought he was too old to run anything, so they had 

Maverick a 49 

consigned him to a minor position in training. We raised him back 

But after a few months in which nothing much happened at the 
plant, Fernando started to lose his patience. Hobart needed drastic 
action, he would plead. Soon he was acting like a German shepherd 
on a chain, ten yards from a collie in heat. 

Relax,” I pleaded. ‘‘Give them a bit more time.” 

But he would just shove in my direction stacks of handwritten 
calculations that proved, he said, the plant was sliding downhill. He 
kept insisting that, although he was our director of administration 
and finance, he still had time to turn Ipiranga around. 

Clovis and I discussed it often and, in the end, agreed with 
Fernando. We were worried about his hard-edged style and lack of 
experience as a general manager. But Fernando had the aggressive 

personality that we then believed a successful business required, 
especially one in a slump. 

One sunny afternoon we pointed him in the direction of Ipiranga, 

let him off the leash, and watched him bound forward, barking all 
the way. 



Fernando was convinced the Hobart plant lacked organization, 
ambition, and controls, and proceeded to supply these missing 
elements singlehandedly. He would arrive each day at 7:30 a.m., to 
find himself alone in the office until about 9:00, when people would 
finally start filtering in, have their coffee, read their newspapers, and 
only then get to work. And at 5:30 they left for home, leaving 
Fernando to labor until 9, 10, sometimes even 11 p.m. This didn’t 
please him, and soon everyone knew it. 

1 wasn t as driven as Fernando, but it was close. He and I would 
often arrange to meet at 8 p.m. and end up leaving the building 
together after midnight. The difference between us was that I didn’t 
expect everyone else to stay late. 1 may have been a bit irritated 
when, at 6 p.m., I was told that someone I needed to talk to had 
just left, but I also was surprised when I found a colleague around at 
8 p.m. Fernando demanded it. 

lotti, Clovis, and many others at Semco worked long hours, too, 
and their families were beginning to complain. Iotti lived an hour 
from the plant and would arrive home almost every night to find his 
family asleep. Then he would leave the next morning before they 
awoke. Looking back at that time, it was a wonder no families broke 
up. I guess it was mostly because we all were convinced the long 
hours were temporary and that normalcy would return as soon as 
we digested our acquisitions and came to terms with Semco’s 

52 A Ricardo Semler 

100-percent-a-year growth rate. It took us almost a decade to learn 
that our stress was internally generated, the result of an immature 
organization and infantile goals. 

Fernando changed everything about the Hobart plant in his first 
few months. Sales reps and other employees were fired for lack of 
motivation or competence, products were overhauled, prices were 
changed. Some production machines were discarded, others acquired. 
The plant was in perpetual motion. But I would soon have cause to 
wonder whether all this movement was taking it in the right 

At the Hobart plant, and all over the new, improved, giant-sized 
Semco, we could track with great precision virtually every aspect of 
our business, from sales quotes to the maintenance records of each 
of our innumerable welding machines. We could generate all sorts of 
reports almost instantly, with dazzling charts and graphs. We were so 
impressed with our statistical abilities that it took us a while to 
realize that all those numbers weren’t doing us much good. We 
thought we were more organized, more professional, more disci¬ 
plined, more efficient. So, we asked ourselves with a shudder, how 
come our deliveries were constantly late? 

Alost of our managers were like Fernando—proponents of classic, 
authoritarian solutions such as rigid controls and long, grueling 
hours. But a few of us were starting to doubt the effectiveness of this 
approach. I was particularly distressed by the malaise that was all too 

apparent in our factories, both old and new. Workers just didn’t seem 
to care. 

As I thought about it, I realized the tough guys had taken over. 
And while I initially liked the idea of a disciplined, hard-driven 
company, run by aggressive managers armed with innumerable 
statistics, 1 was starting to have second thoughts. Work hard or get 
fired. That was the ethic of the new Semco. People were being 

pushed forward. But how much better to have a self-propelled work 

During this time I often thought of a business parable I had heard. 
Three stone cutters were asked about their jobs. The first said he was 
paid to cut stones. The second replied that he used special techniques 
to shape stones in an exceptional way, and proceeded to demonstrate 

Maverick a 53 

his skills. The third stone cutter just smiled and said: “I build 

As I walked around Semco's plants, I had the sense we had far 
more stone cutters than craftsmen. What I wanted, of course, was a 

company filled with cathedral builders, and there were hardly any of 
them at all. 

if anything, the split between the tough guys and the more 
sympathetic souls widened over the next few months. I kept hoping 
we would find some middle ground and Semco would at last slide 
into gear and start accelerating. Then an incident at our annual 
convention brought me up short. Every year our top forty-five 
corporate officers and their spouses would fly off to a luxurious resort, 
where we would spend four days recharging and, well, luxuriating. 
It was the third and last day of one of these junkets. As the managers 
filed into the conference room after lunch, they saw the cartoon on 
the stand where the flip charts were displayed. It pictured an 
overweight Grim Reaper bearing a scythe and drawing blood from 
dozens of little people, cowering in corners. There was no doubt 
about his identity. Was Fernando that overbearing and tyrannical? 
Was he draining the company’s lifeblood with his autocratic ways? 

“Dick, can I see you a moment?” said Renato Bernhoeft, a 
consultant in organizational behavior who put together the convention. 

This looks like trouble, Renato,” I said after we adjourned to the 
privacy of the hall. 

You bet, Renato said, puffing his pipe. “But it might be the 
opportunity you need.” 

I don t understand. The first two days of the conference went 
well—no open disagreements, no hostility.” 

“Open, that’s the key word. You have two cliques here, Dick, and 
they kept to themselves. There was no mingling between them.” 
Renato took another puff. But let’s be frank about it. These guys are 
locked in battle. The hard against the soft. The rightists against the 
leftists. Renato paused and puffed some more, the way pipe smokers 
do when they want to make a point. “You’ve got to start dealing 
with this conflict. Semco has to have its own culture. A single 
culture. Right now, it’s a mess.” 

54 A Ricardo Semler 

“So what do I do?” 

Have you ever seen a group therapy session?” 

With a group of forty-five angry managers? No way.” 

I dont think you have a choice,” Renato said, and then told me 
what he had in mind. 

What followed can only be described as terrifying. I went back 
into the conference room, where the managers were sitting in stony 
silence. Some were smoking in the nonsmoking section, a clear sign 
of hostility. I walked to the front of the room and tried desperately to 
remember passages from the psychology manuals of Jung and Laing, 
even as I realized that Reich was going to be of more help here. After 
all, he was the inventor of physical psychotherapy, and throwing 
things at one another seemed appropriate. 

Could you all please pick up your chairs and move them 

“Why?” two or three managers shot back at once. 

“Just, please, do it,” 1 said, trying to be firm. “Put your chairs in a 
semicircle right in front.” 

1 wanted the audience to be closer to me and one another. I 
noticed, though, that they were moving carefully, placing their 
chairs near people they sided with. 

When rearranging was done, 1 told them it was time for a candid 
discussion of what was bothering us. I talked for a bit about how we 
all needed to work together, harmoniously, using all the usual 
phrases. At least Renato was nodding. 

“Just a minute,” Fernando bellowed. “This kind of talk is cheap. 
Sure, we’d all like to get along. But first there has to be respect.” 

“Well, no one will disagree with that,” 1 ventured, in my best 



"What I mean," Fernando demanded, “is someone will have to 

apologize for that”—he pointed to the cartoon—“before any conver- 
sation starts.” 

Silence all around. 

“Fernando, surely you can see it wouldn’t be productive to go in 
that direction,” 1 pleaded. 

“It’s even worse to accept that cartoon as a basis for conversation.” 

Maverick a 55 

"Let’s just forget the cartoon, Fernando. It’s just someone’s way of 
expressing his feelings." 

Oh, yeah? What if I decided to express my feelings by punching 
whoever drew it in the mouth?” 

Renato stopped nodding. 

“You’d never do that, would you, Fernando?” I said, switching to 
my kindergarten teacher voice. 

‘‘That’s what he deserves.” 

Just then Oswaldo Guimaraes, our engineering manager, got up. 
Everyone knew he had drawn the cartoon. He was known for his 
cartoons, though they were usually funny, not bitter. But he drew 
this one because he thought Fernando had been much too heavy- 
handed, and it was clear others in the room agreed with him. 

Before he could own up to the artwork I waved at him to sit 
down, saying it didn’t matter whose drawing it was. The issue was 
out in the open and it was time to deal with it. 

Fours hours later, all had been said, but there was still a deep split 
between those who believed in law, order, and organization above all 
and those who felt that people, motivated by a sense of involvement, 
could overcome any obstacle. It was a case of what, in German, is 
called Weltanschauung how you see the world. The autocrats at 
Semco were convinced that nothing would get done if they didn't do 
it themselves or push their subordinates into doing it. They viewed 
company tasks much as parents see homework—disagreeable per¬ 
haps, but mandatory. The touchie-feelie crowd, on the other hand, 
was confident there was a better way, and that it involved giving up 
their power. It was Waterloo versus Woodstock, and I didn’t see how 
our Napoleons and our Timothy Learys would ever be reconciled. 

As we concluded the convention and prepared to go home I knew I 
would have to act. The all-together-now feeling of the Gabriele- 
Harro years was long gone, and the long hours, which then seemed 
joyful and ecstatic, were now tedious. Semco was divided and 
confused to the point of paralysis. Even small problems were difficult 
to resolve; big ones were impossible. 

Studying classic business cases, 1 discovered we were going through 
the ‘‘Bureaucracy Phase” or ‘‘Adolescent Phase,” depending on which 
author I consulted. This was a period in which a business that has 

56 A Ricardo Semler 

expanded starts to suffer growing pains. In response, Fernando had tried 
to cut through the malaise by introducing time-tested techniques of 
discipline and control. His goal was to instill a sense of organization, 
to know what was going on, or at least have the impression he knew 
what was going on. The executives who closed ranks behind him 
didn t mind if our employees came to work scared or anxious, as long 
as they got things done. 

As a result, some of us who felt differently found ourselves trying 
to manage wave after wave of personnel problems and discontent. 
How could I stop it? 

While my mind wrestled with this conflict, my body convinced 
me it was time for a change. 

I was visiting a pump factory in Baldwinsville, New York, when I 
suddenly felt ill and again passed out on the shop floor. I was taken 
to a physician in town and rested in his office for a couple of hours. 
He was an old-fashioned family doctor, the type who likes to talk 
with their patients. He was quite worried about me and advised me 
not to continue my trip without stopping for tests. He suggested I 
might visit the Mayo Clinic in Rochester, Minnesota, or the Lahey 

Clinic in Boston. I was headed toward Boston anyway, so Lahey it 

1 called ahead and talked to an internist there, who set up my 

So, what is it you’re feeling?” asked the gray-haired doctor when 
we met three days later. 

”1 have a constant throat infection that doesn't respond to antibiot¬ 
ics anymore,” 1 began. "Because of this, I can't eat solid food before 
noon, so 1 have chronic anemia. 1 also suffer fainting spells once in a 
while, as you know. 1 take medicine for an accelerated heartbeat and 
for migraine headaches. I have gastritis and drink milk several times 

a day because of heartburn. Oh yes, and I get rashes on my back 
whenever I’m tense.” 

1 watched him scribble all this down. 

‘‘Do you play any sports? Jog? Walk?” 

“Only up the stairs to my office, and then I have to grab the wall 
for a few minutes.” 

Maverick a 57 

I saw the faint trace of a smile as he scribbled. 

“How many hours a day do you work?” 

“Ten. Or twelve.” 

“Sometimes more?” 

“Well, yeah. Sometimes I come in at 7:30 and leave after 

“That’s more than ten or twelve hours.” 

Yes, but that’s not every day. Mostly I leave around 9 00 or 
10 : 00 .” 

“Do you work on weekends?” 

‘A couple of hours on Saturday, and another few on Sunday.” 1 
thought about the exasperated look I got from friends when, heading 

oft to the beach, I’d first fill the trunk with reams of telexes, 
memoranda, and trade magazines. 

“What do you eat?” 

I cant have breakfast, because of my throat, so I just have orange 
juice. Then a sandwich at my desk for lunch. For dinner, just 

whatever’s in the fridge—I’m so tired by then that I don’t have much 
of an appetite.” 

The doctor finished writing and asked me to move into an 

adjoining room, remove my clothes, and sit on the examination 

table. A laint smell of ether came through the door, and 1 felt a little 

You re becoming pale, Ricardo, he said as he put the stethoscope 
on my heart. 

The touch of the cold metal did it. I had a sensation of flying in 

slow motion over a mountain village. All those tiny homes_they 

looked just like dollhouses. “Breathe deeply, breathe deeply,” I heard 
the doctor say as he slowly came back into focus. 

I came to know the corridors of that hospital well during the three 
days 1 spent amortizing every machine the doctor had. I also passed 
out again, when he injected contrast dye into a vein to take X rays of 
internal organs too numerous to list. No part of me was left 

Finally it was time for a reckoning. The kind doctor had a stack of 
yellow envelopes in front of him containing my test results. After 
years of telling my woes to physicians who never came up with a 

58 A Ricardo Semler 

diagnosis, I believed I would finally learn what was wrong. I was 
nervous, but also relieved. I was sure that he would want me to stay 
at the clinic for a couple of days—maybe a week or two—getting all 
kinds of shots and treatments. I was prepared for that, and would be 
phoning my parents and the company shortly to let them know that 
I d be away for a while. I would buy some novels, and relax. If my 
first vacation in years was going to be spent in a white gown, so be 

it. I watched as the doctor put the last X ray in its folder and looked 
at me. 

“You have nothing whatsoever, Ricardo.” 

Excuse me, I thought, staring back at him. Had he mixed up my 
records with someone else’s? 

“Are you sure?” I managed to utter. 

Quite sure, he said. Your tests are all perfect. But you are 
suffering from an advanced case of stress. The most advanced case I 
have ever seen in a person of twenty-five.” 

What happens now?” 

You have two choices. Either you continue your current life, in 
which case you will be back with us, or else you must change.” 

"Change? How?” 

"That’s not for me to say. All I will say is that absolutely 
everything about your life has to change. I’ll recommend that you 
take two aspirins, eight times a day. These sixteen pills won't do very 

much, except remind you regularly that you have a huge problem to 

Then he showed me the way to the cashier. 

We simply do not believe our employees have an interest in coming in 
late, leaving early, and doing as little as possible for as much money as 
their union can wheedle out of us. After all, these same people raise 
children, join the PTA, elect mayors, governors, senators, and presidents, 
they are adults. At Semco, we treat them like adults. We trust them. We 
don t make our employees ask permission to go to the bathroom, or have 
security guards search them as they leave for the day. We get out of their 
way and let them do their jobs. 




I took the sixteen aspirins for just one day. Then I started making 

Before I could reorganize Semco, I had to reorganize myself. Long 
hours were the first issue I tackled. They were one of the biggest 
symptoms of time sickness, a disease that afflicts far too many 
executives. So I set 7 p.m. as the time I would leave the office, no 
matter what. After that, I would go to the movies, read books (but 
not business books) anything but work. I wouldn’t do any work on 
the weekends, either. And onto each long business trip I would add a 
week of pleasure travel. 

Next, I resolved to delegate furiously, and to summon up the 
courage to throw unneeded papers away, so they wouldn’t clutter my 

desk or thoughts. I would follow my intuition more and listen less to 

Yes, I would attack my problem directly, and that problem was 
not simply the management of a business but something even more 
fundamental: the management of time. So many executives find that 
the daily quota of twenty-four hours is too few to get to everything 
and still have any left over. I thought long and hard about time in 
the weeks after my visit to the Lahey Clinic. I realized that if I was 

going to find a cure for time sickness I first had to identify its 

# # # 

62 A Ricardo Semler 

Cause l. The belief that effort and result are directly proportional. 

Order and Progress, the Brazilian flag proclaims. “Order or 
Progress is more like it, since they usually aren’t found together. In 
business, effort is too often confused with result. The sales manager 
who overflows with charm when talking to customers and, after 
closing a sale, takes the rest of the day off to celebrate is regarded as 
lazy but lucky, not as a talented sales executive. 

WBen asked the reason ior their success, entrepreneurs are fond of 
saying, “A lot of hard work.’’ Sounds good, doesn’t it? It plays well at 
home, too, to families who have been ignored for years. But if great 
entrepreneurs were to answer the question honestly, most would 
probably list such factors as a finely tuned sense of timing, the 
ability to recognize opportunity, friends in the right places, an 
occasional moral lapse, and luck. May Horatio Alger forgive me' but 
hard work by itself is not enough. To say it is possible to establish a 
successful business just by arriving early and staying late is like 
saying that every mailman can be Howard Hughes. There is a 
prevailing conviction that sweat is obligatory, and with each new 
drop an executive moves a littler closer to financial heaven. I had to 
rid myself of this notion. It isn’t healthy. It isn’t even true. 

Cause 2. The gospel that the quantity of ivork is more important than the 
quality of work. 

This is a variation on the same theme. Early in the century Max 
Weber recognized that the Protestant ethic of hard work had perme¬ 
ated the business world. It is even more of a factor today. Executives 
feel pressure from their bosses to outwork colleagues and build their 
image and career. By this reasoning, having a heart attack because of 
work leads to true glory and keeling over at the office is even 
better a sign, a Calvinist might say, of being among the Elect. 

Someone who manages his time is often suspect. And if he goes to 
the theater, doesn’t carry a briefcase home, spends weekends with his 
family, and sometimes even picks up the kids at school for lunch, 
then he has already descended into an advanced state of indolence. 

The executive who judges his contribution in hours will find 
himself muttering things like: “Well, we all know how unfair it was 
that they didn’t promote me. Everyone knows I’m here at 8:00 in the 

Maverick a 63 

morning and 8:00 in the evening.” Or: “My daughter needs to make 
an appointment to speak to me.” 

Cause 3. “Things are a little uncertain at the office right now. Ill just 
have to work a little longer until they straighten out.” 

Few excuses are as convincing as the “were just going through 
^ ~• Fill in the blank, a switch at the top, restructuring, 

layoff, expansion. Almost any change can be an excuse for poor time 
management. To allow such events to shape one’s workday is to 
become a mere cork that bobs up and down on the sea. 

Cause 4. Fear of delegation , and its cousin, fear of replaceability. 

Here is where we lay bare some nerves. Fear of delegation is the 
belief that no one is as competent to solve a problem as you are. This 
kind of thinking (which at times may be justified) usually results 
from the belief that tasks will inevitably be done poorly if not done 
by capable hands—yours, of course. But how often is this really 
masking the fear that others can perform jobs you once thought only 
you could accomplish? 

This in turn leads to the fear of replaceability. This means 
postponing vacations, or taking them but leaving phone numbers 
where you can be reached morning, noon, and night—and then 
being disappointed that no one needed you while you were gone. 

I’ll have more thoughts about time sickness and how to cure it later. 
For now, let me say that I have recovered so completely that I no longer 
wear a wrist watch. I gave it up soon after attending a concert by 
Brazil’s most famous pianist, Madalena Tagliaferro. As I listened to her 
play Sibelius, I realized that she had been born when Brazil was a 
monarchy, witnessed the invention of the automobile and the airplane, 
lived through two World Wars, and was still performing. It struck me 
that time should be measured in years and decades, not minutes and 
hours. It is impossible to understand life in all its hugeness and 
complexity if one is constantly consulting a minute counter. 

But how could 1 spread such an idea through a company being run 
as if every millisecond would dent our balance sheet? I couldn’t. 

64 A Ricardo Semler 

There was no way around it. Fernando would have to go. I admired 
his aggressiveness, energy, and dedication, but he needed a harsh, 
intimidating environment for his magic to work. If it was magic. 
After my visit to the Lahey Clinic, I was pretty sure it wasn’t. 

As soon as Fernando parted company, we began to realize that the 
bookkeeping system he and Ernesto had installed was actually 
hurting us. For starters, we now had an accounting department full 
of people who only stopped cranking out numbers to pick up their 
paychecks. And we had so damn many numbers, inside so damn 
many folders, that almost no one was looking at them. But no one 
would admit it. Everyone just bluffed their way through meetings, 
pretending to be familiar with every little detail. In retrospect, we 
realized that it was during this time that we knew the least about 

what was going on at Semco. It was a classic forest-versus-trees 

So we simplified the budget system, slashing the number of “cost 

centers from 400 to 50 and beheading hundreds of accounting 

classifications and dozens of budget lines. We also reduced the 

number of documents we circulated, and the signatures necessary to 

approve them. Our new system was simpler, with limited but 
relevant data. 

As for our actual spending plans, we would eventually develop two 
budgets: a five-year plan and a six-month report. Yes, I know the 
argument against five-year plans. The Soviets used them, and look 
how that turned out. But when we look five years forward, we can 
ask ourselves whether we want to be in a particular market, or 
whether we should drop a product, or whether we will need a new 
factory, among other such questions. So a five-year view is essential. 

In contrast, we take an operational view of six months, because we 
found that in a conventional one-year plan people will invariably 
believe that conditions will improve just enough to compensate for 

the problems they know they’ll have in the first half of the year. Or 
vice versa. 

In either plan we try to think in "zero-based” terms. Budgets 
should always be based on rethinking the company; most of the time, 
though, they re not much more than last year’s numbers projected 

Maverick a 65 

forward and are about as good as warmed-up coffee at two in the 

My constant insistence on seeing only the big numbers is the 
subject of interminable jokes at the company. The finance people say 
they only arrive at the big numbers by adding up all the little ones. 
Therefore, they go on, a budget with only the big numbers actually 
requires more effort to understand than one with every little detail. 
This is an expensive fallacy, but one that is difficult to eradicate. 

One more word on financial planning while I’m on the subject. 
The receipt of cold numbers, even when they are current and correct, 
is not enough. In addition to comparing the numbers that arrive 
every month with the budget, it is essential that they be compared 
with the expectations of the person who is going to use them. At 
Semco, we introduced a program that requires each executive to 
make an educated guess about the revenues, expenses, and profits for 
his department at the end of each month. A few days later, the 
official report is distributed. The comparison gives everyone a sense 
of how much each manager actually knows about his area. 

Simplifying our budget process didn t solve all our problems. But it 
did help us see them more clearly. 

We were still late on deliveries. Authorization forms still spent 
days bouncing from one person to another; people were afraid to sign 
their name to anything, for fear they would stand out in the next 
accounting report. Worst of all, Semco was a company full of 
fiefdoms: each department defended its turf at all costs. Salespeople 
thought the people in marketing lived on the moon. Marketers 
thought the salespeople could see all the way to their navels. 
Financial types believed the plant managers wanted to fill the 
stockrooms to the ceiling with inventory. Purchasing thought the 
people in administration hadn’t a clue about how irritated suppliers 
get when they have to wait for seven signatures before they get their 
check. And production thought the people in sales imagined they 
were living in Japan, with overnight delivery dates. 

My colleagues and I tried various quick fixes for our foundering 
organization, from suggestion boxes to leadership training to 
Japanese-style Quality Circles, in which a range of people from the 

66 A Ricardo Semler 

same business unit, from janitors to executives, would sit down 
together to find solutions to common problems. I tried all the 
prepackaged ideas I could find, scouring every damn business book 
with a title that began with a “How to. . or ended with the word 

System or Method.” I read Alfred Sloan's My Years with General 

Motors and Tom Peters's In Search of Excellence. I studied popular 

business writers such as John Naisbitt, Peter Drucker, Alvin Toffler, 

and Robert Townsend, and academic ones, such as Kenichi Ohmae,' 

Marvin Minsky, business writer Ichak Adizes, and Henry Mintzberg! 

I had two bookcases full of business books, each floor to ceiling. I 

also sought out executives at other companies, grilling them about 

management styles over lunch. I picked up plenty of ideas and 

techniques, but I just couldn't make them work in our offices or 

factories. Our people would be motivated for a while, but then slip 

back. I began to suspect that Semco’s problems went deeper than I 
had realized. 

When I first started working at Semco, I was shocked by the 
oppression I found. To someone who spent much of his adolescence 
playing rock n roll, the rules and procedures at Semco left me 
numb. Still, I tried to become the embodiment of a traditional 
executive and had, I thought, succeeded. I wore the right suits. I 
thought the right thoughts. I was so proud when I devised a 
tardiness program that docked an employee’s wages if he was just a 
few minutes late. I was a tough guy. 

Now, I had a different view. Semco appeared highly organized and 
well disciplined, and we still could not get our people to perform as 
we wanted, or be happy with their jobs. There weren't enough 
cathedral builders. If only I could break the structure apart a bit, I 
thought to myself, I might see what was alienating so many of our 
people. I couldn t help thinking that Semco could be run differently 
without counting everything, without regulating everyone, without 
keeping track of whether people were late, without all those numbers 
and all those rules. What if we could strip away all the artificial 
nonsense, all the managerial mumbo jumbo? What if we could run 

the business in a simpler way, a more natural way. A natural 
business, that’s what I wanted. 

The more I thought about it, the more I was convinced the whole 

Maverick a 67 

company needed to change. I had no grand plan. There was no 
blinding revelation. Just a sense that there was a lifelessness, a lack 
of enthusiasm, a malaise at Semco, and that I had to change it. 
People weren’t gratified by their jobs and often seemed oppressed by 
them. The traditional attitude about workers was that you couldn’t 
trust them. You needed systems to control them. Yet at Semco the 
system was dispiriting and demotivating them. So, I thought to 
myself, why not start by eliminating some of the most visible 
symbols of corporate oppression. 

I had always wondered about the security check our people were 
forced to submit to on their way out through the plant gate. It 
bothered me that we treated even veteran workers that way. I asked a 
top executive about it in my first year at the company. 

“Everyone does it,’’ he patronizingly assured me. “Theft is so 
common we need to check everyone, every day. We can’t make any 

He made me feel so naive for thinking that business could be 
conducted based on trust that I didn’t bother to ask him about the 
time clocks or the invariably burly watchmen who loomed over 
them, making sure incoming workers didn’t cheat by punching 
anyone else in. 

Nearly every company of any size has its own FBI. Some even have 
their own J. Edgar Hoovers. Yet these same companies tell their 
employees they’re all part of one big, happy family. How can they 
rationalize such sanctimonious sentiments when they frisk their 
workers on the way home? Or deduct vacation time when someone 
arrives ten minutes late? Or audit the petty cash account of someone 
who has been with the company for two decades? Or put padlocks on 
the storerooms to prevent the entry of “unauthorized personnel’’? 
What family searches its members for silverware as they leave the 
dinner table? 

Workers are adults, but once they walk through the plant gate 
companies transform them into children, forcing them to wear 
identification badges, stand in line for lunch, ask the foreman for 
permission to go to the bathroom, bring in a doctor’s note when they 

68 ▲ Ricardo Semler 

have been ill, and blindly follow instructions without asking any 
questions. 7 

So I decided to end the searches at Semco. It wasn’t hard. I just 
had a sign posted at the gate that read, please make sure as you 


I knew some managers would be appalled. What 1 didn’t anticipate 
was that shop stewards would complain. Some workers, they said, were 

W0 "' ed ,^ ey WOuld be b]amed if a tool disappeared, as it inevitably 
would. Our people want the searches,” the union leaders pleaded. 

hey want everyone to know that they're not the ones taking the 

C °° S u " after two hand-held drilling machines disappeared from 
the Hobart plant, so many workers demanded that searches be resumed 
we had to hold a plantwide assembly to calm everyone. 

Imagine! Workers wanted to be searched to prove their innocence. 

• e w/? “ eXpkln that they Were forsakin « their right to credibil- 
ity. at appened to innocent until proven guilty? Isolated thefts 
should not be enough to make them give that up 

I wasn’t under the illusion that, by eliminating the searches we 

W f ° UW e ‘ m ' nate thefts ’ 111 bet that on average 2 percent or 3 percent 
o any work force will take advantage of an employer’s trust. But is 
this a valid reason to subject 97 percent to a daily ritual of 
humiliation? Yes, there will be theft here and embezzlement there 
ut thats the case in companies with huge auditing and monitoring 
departments. It s a cost of doing business. I would rather have a few 

Li'strusT^ m 3 Whlie th3n C ° ndemn eVery ° ne C ° 3 SyStem based 

While I was at it, I took out the big time clock at the plant cate 
and had small clocks installed throughout the plant and offices Then 
we suggested to everyone that punching in for their colleagues would 
not be appreciated, and let it go at that. Today, the time clocks 
adhere only to help our employees keep track of the hours they 

Have thefts and time card cheating increased or decreased > I don't 
know and I don't care. It’s not worth it to me to have a company at 
which you don t trust the people with whom you work. 7 

Maverick a 69 

Nothing seems more medieval than dress codes. Office personnel 
are supposed to stroll around in suits and ties or dresses, but who 
remembers why? It’s like this because . . . well, it just is. 

The receptionist is the calling card of the company—how silly is 
that? What customer, supplier, or banker would cancel a deal when he 
sees a casually attired receptionist? What buyer has failed to do business 
with a company because a salesman wasn’t sufficiently fashionable? 

Dress codes are all about conformity. People want to feel secure, 
and dressing like everyone else is one way to accomplish it. If 
everyone at IBM wears blue suits and white shirts, then even a 
trainee will feel he is part of the company if he is so attired. But the 
flip side is that these same people will come to depend on other 
forms of artificially imposed unity, such as a uniform language, 
uniform behavior, maybe even uniform thinking. At its worst, a 
company turns Orwellian and creativity and freedom are smothered 
by discipline and the weight of shared expectations. Obviously, few 
corporations let it go that far, but even relatively informal businesses 
have unwritten codes of conduct. The more elaborate they are, the 
worse it is for flexibility and, ultimately, profits. 

Of course, some people say they like suits and dresses. Clovis, for 
example, always comes to work in a suit and tie. Harro, on the other 
hand, got a roomful of hoots when he pulled a suit from the depths 
of his closet and wore it to an important meeting. If I needed any 
proof of the ridiculousness of dress codes, I had only to recall how 
our old friend Ernesto Gabriele brought his old ties from home for 
the clerks in the accounting department, who didn’t have any of their 
own. It was one more example of how Ernesto tried to create an 
atmosphere worthy of a multinational, but with their knots tied 
wrong and their shirttails hanging outside their pants, the clerks 
looked like they were headed to a Halloween party. 

Why is it that when they come to work on weekends, people 
invariably dress in casual clothes? Because they feel more comforta¬ 
ble. Well, why shouldn’t they feel more comfortable every day? So we 
told our office workers and managers they could dress as they 
pleased. Period. And most gave up suits and ties and dresses in favor 
of jeans and, on hot days, shorts. Sure, there are moments when more 
formal attire is appropriate, such as at board meetings and presenta- 

70 A Ricardo Semler 

tions to important customers. But every responsible adult knows how 

to dress correctly for these occasions. In fact, one of the few 

complications of our new policy occurred when one of our air 

conditioner repairmen was denied entrance to the Citibank Building 

in Sao Paulo because he was dressed too casually. To fix an air 

conditioner! He went home, changed, and went back to work. End 
of problem. 

• W t h °P ed eliminat >ng the dress code w ould help create a company 
in which office doors would seldom be closed and it would be 

common for people to walk in, sit on a colleague's desk, and 

eavesdrop on a meeting that had nothing to do with him. Today I am 

a lg believer in MBWA, or Management by Wandering Around 

Popularized at Hewlett-Packard, it simply means taking time each 

week to walk around with, as Bob Dylan said, no destination known 

You can see how new projects are going, solve some problem m the 

actoiy or t e o ice, or just chat in the hall with someone you haven't 
seen in a Jong while. 

S °M J th °“ gh ’ y ° U Can ' C bfeak down the walls until you 
actually break down the walls. So one day 1 called in everyone on the 

third floor, one by one, and asked them what they would think about 

giving up their individual offices and sharing one big, beautiful work 

space divided by plants and flowers. We formed a committee and 

soon enough the wrecking crews arrived. My goal was for each person 

to have a desk wherever he wanted, which made it all rather 

finTsh ma monl ^ ^ ^ “ - ™^d to 

c . 11 s “ much —^ 1 

Suits and ties were easy compared to workers' uniforms. 1 remem 
er a meeting of top managers at which several major investments 
were approved in a matter of minutes. Then we came to the matter 
of replacing our factory workers' sky blue work outfits. We seen 
more than an hour arguing about the relative merits of various 
hues which was more sober, which more inspiring, which showed 
dirt less without reaching a conclusion. Everyone had a chance to 
express an opinion on the uniforms, I thought to myself, except the 

Maverick a 71 

people whose opinions should count the most: the people who would 
wear them. 

Clovis, as usual, was reading my mind. “What if we took a survey 
of the workers to see what color they prefer?” he asked. 

“Are you crazy?” was the response from around the table. “They’re 
going to want yellow, orange, white—it’ll be a nightmare.” 

What could be expected of workers? the managers argued. Cer¬ 
tainly not an understanding of the influence of colors on motivation 
and productivity. Who could expect them to be practical? 

I couldn’t believe what I was hearing. I mean, we were talking 
about the color of clothing. In keeping with our new thinking— 
Clovis’s and my new thinking, anyway—we put the survey idea to a 
vote and the nay sayers around the table reluctantly agreed, thinking 
no doubt that the workers would want impractical colors and they 
would have an opportunity to tell the boss “I told you so.” 

We polled workers in three of our four factories, offering them a 
choice of all the colors of the rainbow and a few more commonly 
associated with controlled substances. The result was a tie between 
light blue and a blue-gray called petroleum blue. We had to take a 
second survey, which was accompanied by the sneers of the skeptical 
directors. “By now everything could have been purchased and delivered,” 
they didn’t have to say. 

This time petroleum blue won by a landslide. Even the critics 
reluctantly conceded that it was more practical than our old sky blue, 
because grease stains wouldn’t be as visible. We told ourselves the 
hours we spent distributing and tabulating the survey twice was 
worth it. 

Emboldened, I took on a really serious problem. At Semco, one of 
our biggest headaches was parking. We didn’t have nearly enough 
spaces at our Sao Paulo headquarters. After much discussion among 
managers, we solved the problem democratically: we abolished 
reserved spaces for big shots and apportioned them by department, in 
proportion to their size. Directors and managers ended up with eight 
spaces, administration had ten, clerks and other office workers ten, 
machine operators fifteen, and so on. Within each department, all 
spaces were available on a first-come, first-served basis. Only after a 

72 A Ricardo Semler 

few executives were forced to leave their cars outside the gates and 
saw to their astonishment that their instructions continued to be 
carried out by the workers who had parked inside the gates did they 

acknowledge that respect is not a function of the distance from car 
door to plant door. 

In the next few months, Semco asked executives to share secretar¬ 
ies (about whom much more later); eliminated private dining rooms; 
and allowed managers of the same rank to have different kinds of 
desks and chairs and occupy offices of varying sizes. No longer would 
it be possible to discern a person’s status by the grade and grain of 
his office furniture or the plushness of his carpet. 

Democracy is a lot of work, I kept telling myself and anyone who 
would listen. It needs to be exercised with conviction and without 
subterfuge or exception. And it begins with little things, like 
neckties, time clocks, parking spaces, and petroleum blue uniforms. 



An incident not long after Fernando arrived at Semco should have 
warned me that he and 1 were destined for trouble. I had talked him 
into accompanying me to the headquarters of the Metalworkers’ 
Union of Sao Paulo for a bargaining session. The trip was unusual 
because in Brazil companies always make the unions come to them. 

There was no mistaking the union office. Dozens of large, stocky 
workers, dressed in overalls and blue jeans, lined both sides of the 
corridor. How were we going to get past all those angry-looking men? 

Fernando didn’t slow down, so neither did I. As we got closer, the 
workers looked even more menacing. Some moved away from the 
walls they had been leaning against and stood straight up. A discreet 
entrance was out of the question. Not only were Fernando and I both 
more than six feet tall, but we were also dressed in gray flannel suits, 
with sober striped ties. 

The workers started to murmur. “It’s them,’’ one of them hissed. 

We walked forward until we were surrounded. 

“It’s them, the bosses,’’ said another. 

Threatening looks abounded. Violence did not seem out of the 
question. Just as I was about to try to explain who we were and what 
we wanted, the union chief appeared. 

“Relax, guys,” he said. “These two aren’t from Taurus.” 

Taurus, a weapons manufacturer, had just Fired the workers at the 

74 A Ricardo Semler 

union hall for striking. They were expecting a visit by representatives 
from management and assumed that's who we were. 

The workers seemed more amiable and Fernando and 1 recommenced 
breathing. Some even smiled at us. I smiled back with great 
enthusiasm. But Fernando walked on, fists still clenched. 

The union leaders, not used to being hosts in such circumstances, 
told us our visit was a great honor. In fact, Walter Schiavon, the 
union director for our region, explained that most companies did 
everything they could to demean and intimidate the union leaders, 
starting with keeping them waiting when they arrived to bargain. 

They make us feel unimportant, as if the managers are so busy 

they can barely fit us into their schedules," Walter said. “We become 

impatient and irritated, and are less attentive when the negotiations 

“Ah," said Fernando, looking over at me. “We must learn to do 
that, Dickie.” 

I squirmed a bit in my chair. Fernando was half kidding of 

course, but it was that other half that bothered me. I had ^been 

t inking about softening our hard line toward the union. Fernando 
clearly wasn’t with me. 

Thats just the beginning," Walter went on. “They make us sit 
across from the window, so the light shines in our faces. But they are 
in shadow, so it's hard to see their reactions.” 

"And sometimes the chairs we sit on have one leg that is shorter 
than the others, so we don't feel secure," another union man said 

■ °' 0 :; airS 1 ® Sh0rter than theirs > 50 we have to look up at 

them, Walter added. F 

I looked over at Fernando. He was smiling. 

Once everything was handmade and expensive. Then Henry Ford 
set up his assembly line. Industry began to cater to an emerging 
middle class, increasing volume, diminishing costs, and providing 
goods for the multitudes. But if the artisans of the eighteenth and 
nineteenth centuries could perform highly skilled tasks with little or 
no supervision. Ford's factory workers, who had virtually no skills 
required foremen to watch over them. And who were these foremen 
Ut po ished versions of the same workers who also needed supervi- 

Maverick a 75 

sion. And their supervisors needed department heads. And the 
department heads needed vice presidents. So mass production spawned 
a huge bureaucracy. 

The industrialists and their shareholders could hardly have done 
better. You won’t meet anyone named DuPont living under a bridge, 
or need to feel sorry for Henry Ford III, no matter what Lee Iacocca 
says. Little by little, though, laborers started wondering about their 
end of the stick. Factory workers, especially in the United States and 
Europe, began to question the merits of burning themselves out, 
physically and mentally, while toiling in unhealthy environments. 
Unlike artisans, their decision-making power—their influence over 
the conditions they worked in—was negligible. Even in the largest 
businesses, it is rare that more than half a dozen people decide 
corporate strategy and determine the destinies of employees. In the 
face of such a brutal concentration of power, workers feel infinitesimal. 

Tired and deflated by the heat and the smoke and the endless 
monotony of their jobs and depressed by feelings of insignificance, 
workers formed unions that soon enough began to increase wages and 
eventually imposed a straitjacket of rules that sought to give their 
members some control over their lives. The battle was joined and the 
result was an increasing inefficiency. 

In recent years some companies have sought a way out. Mostly 
they are younger firms, less encumbered by traditional thinking. 
They are devising new formulas to make peace with their employees 
and increase their involvement with the company and its goals. I 
wanted to move Semco into their camp. But I knew that we couldn’t 
begin to defuse the hostility and distrust that had built up over the 
years without first improving communication with the shop floor. 
Information was always being filtered through the bureaucracy, which 
was limiting and distorting it. I suspected it was time for a change 
when we hired those time-and-motion specialists to analyze our workers’ 
routines. We thought these experts would help increase our workers’ 
productivity. Much later, the workers told us they had quickly learned 
to slow down the analysts’ timers, which made their study a washout. 

What we needed was a new approach. Despite Fernando’s skepti¬ 
cism, I went ahead and asked our employees at each of our four 
business units to form committees comprised of representatives from 

76 a Ricardo Semler 

every part of the operation but management. Machinists, mechanics, 
office workers, maintenance workers, stockroom personnel, draftsmen— 
every group would have a delegate on these committees, which 
would meet regularly with the top managers at each plant. 

Only a handful of companies in Brazil had factory committees, and 

their members were usually the most obstreperous, uncooperative 

workers on the payroll. Almost all these groups had been shoved 

down the throat of management by unions as a condition for ending a 

strike. A factory committee at a General Motors plant had instigated 

a two-week strike in which it barricaded the main gate with trucks 

held managers hostage inside an office building, burned partially 

assembled cars on the assembly line, and made Molotov cocktails to 
lob at the police. 

We prayed our committees would have more polite ways of dealing 
with us. But we gave them a broad mandate anyway, telling them 
they were to look after the workers’ interests. Members would even 
ave time off, with pay, to dedicate themselves to their new job 
which we fully expected would lead to demands for shorter working 

hours, higher pay, improved working conditions, and maybe even 
better food in our cafeterias. 

We didn’t want the unions to think these committees were 
designed to replace them, so we negotiated their charters with union 
leaders, and gave them a seat on each committee as well 

What we didn’t expect—but should have—was the hostile reac¬ 
tion to our committees from Brazil's business community. The 
ederation of Industries, representing the nation’s industrial aristoc 
racy, was especially peeved. The business magazines attacked us too 
m articles with such headlines as, “These Guys Run Their Boss's Plant.” 

For all the concern, our factory committees were hardly a threat 
At first, they hardly existed. Older, veteran workers didn't want to 
join because they were uncomfortable taking part in what they 
believed was the Boss’s role. They had worked under the traditional 
system for one two, even three decades, always being told what to 
do and how and when to do it, and they didn’t want to change Some 
younger workers weren’t interested in the committees, either Perhans 
they were suspicious of our intentions. Maybe they were scared by the 

Maverick a 77 

increased responsibilities that go with increased participation. Possibly 
they just didnt want to pull up to the same table as our top managers. 

I always hated the bosses,” recalled Demerval Mattos, a lathe 
operator at the marine pumps unit at our Santo Amaro plant. “The 
owners of the company were my mortal enemies. I didn’t want to sit 
down and talk to them.” 

Even so, Demerval joined a factory committee. So did Joao Soares, 
a welder at the dishwasher unit at the Hobart plant. At the first, 
tense meeting, Joao and his colleagues were tongue-tied. “For three 
or four minutes, nobody said anything, nobody knew how to start,” 
Soares would later say. “Everybody was nervous. Some were so 
frightened they never did speak, though the meeting went on for 

For his trouble, Soares was elected head of his committee, which 
hardly increased his enthusiasm. Eventually, though, he and others 
would find these committees one of the most rewarding experiences 
of their work lives. “I saw top management sitting at a table with 
me and listening to what we had to say and willing to do something 
about it,” Soares said. “I saw this was a place where the employees 
could grow and the company could grow.” 

Consider the evolution of Oseas da Silva, another lathe operator at 
the marine pumps unit. “I was,” he says, “a very radical person. 
But by being on the committee, I learned that with dialogue you 
can achieve your goals.” 

When Semco won a national award for labor relations in 1989, I 
decided that, since the factory committees were responsible for the 
honor, a committee member should accept it, even at the risk of 
breaching government protocol. So Oseas da Silva, former radical, 
shared a podium with a president. 

But that was much later. At first, once they found their tongues, 
the committee members concentrated on three issues: money, money, 
and money. Most were reasonable about it, though. They asked us to 
survey other companies to find out what their workers were paid, and 
when Semco lagged behind the industry average the committees gave 
us time to phase in increases. 

Another concern was job security. Committee members didn’t feel 
they could sit across the table from their bosses and speak freely if 

78 A Ricardo Semler 

they could be dismissed for what they said. That seemed reasonable, 
so we guaranteed that they wouldn’t be sacked while they served on a 
committee and for one year afterward. 

That got things rolling. Soon we were being besieged with 

suggestions and demands. One committee produced a list of twenty- 

three items, including transportation to and from work and first-class 

medical insurance. (They eventually got the latter, but not the 

former.) The committees were also curious about the boss’s perks 

“They all had cars,” said Soares, “they all belonged to clubs, they all 
had very nice lives.” 

Our annual convention provoked their scrutiny, too, especially 
when I had to cancel a meeting with a factory committee because it 
conflicted with one of our retreats. “Glad you brought that up,” a 
worker representative said. “Just what is this retreat going to cost?” 

We told the truth—the meetings never cost less than five figures 

"A hell of a waste,” the worker muttered. He wanted to know why 
we had to meet in such a pricey hotel, why our spouses had to come, 
and Just what did we do there, anyway? Now we have much cheaper 
conventions, with fewer managers, and much less fun. 

Through years of heady growth, painful retrenchment, wide-ranging 
reorganizations, acquisitions, and unexpected initiatives, these com¬ 
mittees, which seemed so menacing to so many, assumed uncount¬ 
able managerial responsibilities and became vital to Semco’s success 
They started conservatively, creating subcommittees to modernize 
ocker rooms and bathrooms, buying the materials themselves and 
sending us the bills. Some committees elected mayors for the 
factories to care for common spaces such as gardens and reception 
areas, maintain lighting, even change the furniture. Factory commit¬ 
tees organized teams to paint offices and machines and ran “adopt a 
tree programs that forested our machine shops. At Santo Amaro the 
committee asked one of Brazil's foremost artists, Tomie Otahke to 
sketch a new design for the plant’s interior, and workers plan to 

execute her vivid, geometric patterns when business permits the 
added expense for paint. 

Eventually the committees broadened their concerns, butting into 
our business, which was the running of the business. They helped us 

Maverick a 79 

identify surplus managers, casting suspicious eyes toward people 
with titles such as strategic planning manager, data processing 
manager, even marketing manager. They were constantly questioning 
expenses that executives thought were reasonable, from the gardener 
who did our landscaping (and made more in a week than janitors 
made in a month) to the rent on our factory buildings. They took 
over plant cafeterias, set production goals, and suggested major 
changes in products. 

I remember scheduling a “Chat at Lunch” session with the 
committee at Santo Amaro to discuss sales and billing prospects. 
That morning, without telling me, they sent a delegation to our 
sales and purchasing departments for a briefing. When it came time 
for the meeting, they knew more about last-minute developments 
than I did. 

When plants faced hard times, their factory committees would 
take the initiative and lower wages or increase hours, saving money 
and protecting jobs. When layoffs were unavoidable, the committees 
got involved in the sensitive and unfortunate task of deciding who 
would go. Together we tried to be socially just, taking into consider¬ 
ation such factors as a worker’s history with the company, loyalty, 
ability to find a new job, and family commitments. A person with 
seven children or a spouse in the hospital, for example, would have 
an edge over someone fresh out of school with no responsibilities. 
And under a separate procedure the dismissal of any Semco employee 
who had been with us for more than three years or was over fifty 
years old had to be specifically approved by a long list of people. 

But sometimes the committee members complained that in our 
effort to be equitable we dragged out the process, talking too much, 
agonizing too long, and increasing the pain. Perhaps that was an 
unavoidable price for corporate democracy. 



Fernando, the last of our Attilas, was the first key manager to be 
expelled by the new Semco. He was very smart and very capable, but 
unequivocally autocratic. After he left Hobart we decided we couldn’t 
take the risk of hiring someone new to run the unit, which had 
already suffered too much culture shock. 1 was anxious to show that 
improved performance and a touchie-feelie style were not mutually 
exclusive. This was as good a chance as any. 

I redecorated Fernando’s office, replacing his bleak oil paintings 
with modern posters, including one ol a neon ladder that seemed to 
climb one of the walls, and early one Monday morning 1 moved in. 
But the new general manager didn’t spend much time in the office. I 
immediately started wandering around. 

My first stop was the sales office. I wanted to talk to Mara 
Mantovani, our marketing manager. She had fought constantly 
with Fernando and was known to make deprecating remarks about 
the plant’s management. 

Mara looked up, surprised, from her desk. "I was just coming over 
to see you,” she told me. 

“Let’s just meet here,” I said, sitting down. 

She looked around anxiously, as if regretting that she hadn’t 
straightened up before I arrived. 

“Let’s start with you telling me what bothers you most at 

82 a Ricardo Semler 

“Uh, lots of things,” she said, shifting in her chair. 

Tell me the three or four most important ones.” 

Well, for one, I don’t have the autonomy to print brochures, to 
organize trade fairs, to send out mailings. 1 needed Fernando’s 
approval for everything. And that took a long time.” 

As of today, Mara, we’ll be having a weekly meeting at 9 a.m. 
each Monday for all those in leadership positions. Just present your 
proposals and convince the people there. If we can’t agree, we’ll 

vote.” I paused for a moment to let the idea sink in. “So, what other 
problems do you have?” 

“Oh, lots. I sometimes need to tell a sales rep what kind of 
expense account he has from us, and I can’t decide that myself.” 

“That’s the same sort of problem as the First one, isn’t it?” 

Maybe it is. The weekly meeting, right ? ” 


What about staffing. I need another assistant, and I can’t get 

one. Mara raised her eyebrows, hoping I would immediately author¬ 
ize one. 

“The weekly meeting,” I sang out. “The people there will be 
putting together the budget. What else?” 

"I guess everything I can think of will be decided at these 
meetings, right?” 

“You’ve got it. You'll have to develop your lobbying powers, Mara. 

And since I have only one vote, I'm going to have to do a lot of 

lobbying myself.” I gave her a wide smile as I got up to leave. “See 
you in Congress.” 

1 spent the rest of the morning having the same conversation with 
the other key managers at what we had begun to call our food service 
equipment unit. Then I headed back to my office, which I declared 
to be the plant's meeting room, available for everyone. (It soon 
became common for me to arrive at my desk and find someone 
sitting in my chair, using the phone or having a meeting. I would 
just sit on the visitors side and wait until they were finished.) 

The weekly meetings, which started that first day, quickly changed 
the plant's culture. Instead of waging guerrilla wars with the boss for 
a raise or a new assistant or the freedom to make a decision 
managers now marshaled their arguments and presented them to 

Maverick a 83 

their colleagues. The agendas for these sessions were lengthy at first, 
but over time they shrank as everyone began to make more decisions 
themselves. People only brought up issues they were genuinely 
unsure about. Even then, the group would often just throw the 
problem back in the person’s lap. Mara, for instance, was told she 
should publish brochures in whatever colors she wanted, with 
whatever layout she wished, and hire however many people her 
budget allowed to help. 

“But what if I put out a mailing and the sales reps don’t like it?’’ 
she asked. 

“It’s up to you to decide whether you are going to consult them 
before you publish,’’ I told her when she asked me about it. “If you 
feel confident about it, go ahead. If they don’t like it, your rating as 
marketing manager will suffer, and next time you’ll be sure to 
consult them. We’re not against mistakes here. If you’re not making 
some mistakes, you probably aren’t taking enough risks.’’ 

“But what if they don’t like my work?” Mara persisted. 

“One of two things. If sales go up, they will learn to trust 
your judgment.” I paused for emphasis. “If they go down, they 

“And then?” Mara asked. 

“Well, this group puts together the budget, and they’ll only 
include you in it if they think you re a good investment.” 

Looking back, I can’t remember a single decision that I made in 
that period. Which was just as well, for I am at my best when I am 
doing the least. 

I always smile when I hear executives boast about how participative 
they are. “I want everyone to feel involved,” they will say. “So I call 
everybody in, hear what’s on their minds, and only then decide.” 

What people call participative management is usually just consul¬ 
tative management. There’s nothing new to that. Managers have 
been consulting employees for centuries. How progressive do you 
have to be, after all, to ask someone else’s opinion? And to listen to 
that opinion—well, that’s a start. But it’s only when the bosses give 
up decision making and let their employees govern themselves that 
the possibility exists for a business jointly managed by workers and 

84 a Ricardo Semler 

executives. And that is true participative management as opposed to 
merely paying lip service to it. 

Every few weeks the Hobart plant's managers would spend a lunch 
hour talking to the workers, who would gather in the cafeteria, 200 
strong, and talk about anything that was on their minds. No subject 
was taboo—salaries, profits, new products, hiring and firing policies 
were all fair game. Through these sessions we started to convince 
t em that everyone at our food service equipment unit should make 
more decisions and become more involved in the company. Everyone 
could be a cathedral builder. 

The pot soon began to boil and before long the old Hobart plant 
was unrecognizable. Not so much physically, of course, but 
organizationally and, if I may say so, spiritually. Workers who had 
or years—even decades-reported to the plant and promptly turned 
their minds off became full-fledged industrial citizens, making deci¬ 
sions not only about their jobs but also about the products they were 
making and indeed about their company. But at first, the bubbles 

It all seemed to start when a group of women decided to do 
something about the men's locker rooms, which were always dread¬ 
ful. If they could be trained to take better care of their area the 
women believed, the men might be less careless elsewhere around the 
plant. One Friday night after all the men had left, the women stole 
mto the locker room and found that it was even worse than they had 
imagined. Towels underwear, and toilet paper were strewn about 
Dirty clothes spilled out of bins. With felt-tipped pens, the intruders 
p oceeded to tag scattered items with notes. “/ feel out of place here " 
l want to go back to my locker." "When am I going to get a bath’" 
Corny, right? But when the men arrived Monday they quickly got 
the message. From then on the whole plant was much cleaner Soon 
new showers and lockers were installed. Then some of the men 
redesigned an unused production area, turning it into a game room 
t at was used at lunch and after work. They did all the construction 
themselves and invited the managers to a party when it opened 
Soon chaeflera and giboia plants were sprouting between the 
machines on the shop floor, just as shrubbery had earlier appeared in 

Maverick a 85 

our offices. Then some of the assembly line workers decided to paint 
the factory. Each selected a color for the column nearest him, while 
the walls behind each group of five or six workers were painted a 
collectively chosen shade. One veteran manager consulted a book on 
the psychology of colors and warned us that some shades the workers 
favored would darken the plant, making it depressing and possibly 
dangerous. Limit the choices, he advised. Better yet, choose the 
colors for them. It reminded me of the fuss over the uniforms. I told 
him our philosophy was to leave such decisions to those who would 
be living with the consequences. When one group of workers picked 
forest green, our resident colorist shuddered. Another group wanted 
brown. We kept our mouths shut. The plant was closed for an 
afternoon as workers took up brushes and rollers and turned it into a 
plant of many colors, including the industrially unlikely shades of 
pink and magenta. These hues may have looked chaotic, but they 
were symbols of our desire to let our workers control their own 

The office workers watched all this activity with rising jealousy. 
Why shouldn’t they have more say about their walls? No reason at 
all. So one Friday they all arrived in jeans and overalls and by 
lunchtime dozens of cans of paint were spread out on the floor. There 
were boom boxes, too, and singing, while managers and clerks 
plastered the walls with twice as much paint as our expert said was 
necessary. No one who entered the offices could tell the paint job was 
performed by amateurs and no one who leaned against a wall 
escaped a reprimand. 

The plant may have had the worst food in any institution without 
bars. So the employees formed a cafeteria committee and took 
responsibility for selecting a supplier, monitoring quality, and setting 
prices. Jair Pinto, a clerk in accounting who had been with us for 
twenty-five years, was elected its head. Jair was known for his 
perseverance. He would stalk customers who fell behind on their bills 
like a Doberman pinscher greeting an unannounced visitor at midnight. 
Even so, some workers thought he needed a little motivation as far as 
the food was concerned. One day he arrived at the cafeteria to hear 
murmurs of discontent rising from the long dining tables. It had 
something to do with dessert, a substance barely identifiable as 

86 A Ricardo Semler 

pudding. Jair was doubtless wondering what was going on as he 
started his meal. A worker sitting at his table got up, nonchalantly 
dumped his pudding into Jair’s tray, and walked out without saying 
a word. Moments later another worker did the same thing. Then 
another. In ten minutes Jair’s tray was a mountain of pudding. 

The next day, Jair began to weigh the steaks to make sure they 
met the 125 gram standard specified in the contract. He was all over 
the kitchen staff, too. And soon the quality of the food ceased to be a 
source of complaint. 

That didn’t end the problems at the cafeteria, though. Semco’s 
policy was to subsidize 70 percent of the cost of the meals we served. 
But after consulting with Clovis and some others the factory workers 
instituted a “Robin Hood Meal Plan’’ under which employees paid on 
a sliding scale based on income. Managers and engineers were asked 
to pay 95 percent of the cost of their food, for example, while floor 
sweepers paid just 5 percent. 

A few supervisors felt unfairly treated and responded by bringing 
their lunch from home. The rest of the employees decided to do 
nothing, hoping the supervisors would realize it was cheaper to pay 
95 percent than 100 percent, which, after four weeks, they did. 
Today those supervisors defend the Robin Hood plan. They realize 
what it means to their lower-wage colleagues. 

Small changes eventually led to larger ones, as we knew they 
would. From shrubbery and paint and dirty lockers and lunch 
subsidies the workers moved on to concerns central to their jobs and 
their plant. The factory committee spun off groups that studied the 
plant’s products and how the workers made them, looking for ways 
to save time and make improvements. These teams weren’t created by 
Semco; they formed spontaneously, as the bracing winds of democ¬ 
racy swept through the food service equipment unit. 

One group restructured the dishwasher assembly line, changing it 
from a sequential assembly to a batch process in which dishwashers 
are manufactured in twos and threes by teams of workers who do 
many different tasks and spend the time between batches prefabricating 
the components they need for the next group of machines. They also 
came up with a system in which all the parts for the dishwashers 

Maverick a 87 

were stocked in open racks in the middle of the factory. Metal tags, 
green on one side and red on the other, hung on each rack, and the 
workers would flip the tags when they saw it was time to reorder, 
insuring a steady supply. This was a big improvement on the 
traditional assembly line, in which dehumanized workers have no 
role in decisions regarding the production process. They just do their 
minutely choreographed jobs and, if no bolts are available, simply 
stop working (and love it). In our system, workers were involved in 
what they were doing and hollered well in advance when the supply 
of any part was running low. Production wasn’t interrupted unless 
there was a problem at a supplier. 

Another team of employees changed the casing on our meat 
grinder from steel to fiberglass, which made it easier to clean, more 
modern in appearance, and cheaper to make. Another group dreamed 
up a new way to preweld the base on our scale, saving $27 a unit. 
And yet another group scrutinized the slicers for cold cuts and other 
meats. Hobart had been making these sheers for years, based on an 
old design, and their popularity had plummeted to the point that 
only a few dozen a month were being sold. The workers realized that 
a change from cast steel to stainless steel would make the sheer both 
more attractive and more hygienic. But stainless steel shows finger¬ 
prints, and no one thought customers would want to shine their 
machines every few hours. A matte finish was the answer, but an 
industrial engineer on the team calculated that it would take six 
extra production steps and five additional hours of work. That would 
make the slicer too expensive. 

Annoyed at this stumbling block, the group went off to have 
lunch together. But one worker had an idea, stayed behind, and gave 
a slicer a matte finish in just four steps. When his colleagues 
returned, they were amazed to learn that the new method added less 
than an hour to the assembly time. A new slicer was born, and sales 
shot up to several hundred a month. 

The strength of these groups was their diversity. They included 
factory workers, engineers, office clerks, sales reps, and executives. 
None had a formal head; whoever showed the greatest capacity to 
lead got the job, calling meetings and moderating discussions. In 
more than one group, a shop-floor worker guided professionals. 

88 A Ricardo Semler 

Instead of a seniority system, or boxes on an organizational chart that 
guaranteed power, the groups were held together by a natural system 
of collegial respect. 

There are similarities between this system and the Japanese 
approach to organizing manufacturing operations, but also important 
differences. In our groups, younger members didn’t automatically 
submit to their elders. Moreover, once a team decided an issue, it 
stayed decided. There was no approval needed to make a change. 
Then again, there were no special rewards for new ideas. It was a 
spontaneous process. People participated only if they wanted to. 

Soon every section of the Hobart plant boasted scoreboards above 
the shop floor that tracked the workers’ current production against a 
monthly goal they themselves set. Managers didn’t put the score- 
boards up; the workers did. They even had a master scoreboard in the 
cafeteria that showed the whole plant’s daily output, product by product. 

At first setting goals and monitoring production was uncomforta¬ 
ble for some workers, who were accustomed to arguing that if 
management’s quotas weren’t met it was because executives hadn’t 
planned correctly. Now, the workers couldn’t mindlessly complain 
about how the factory was run, since they were helping to run it. 

Once again, the slicing team led the way. Its members decided 
that they were going to make 210 sheers a month. Not 200. Not 
220. They calculated the time and effort and the materials involved 
and concluded that 210 was the right number. 

We watched closely that first month, curious to see how they 
would do. They started strongly, averaging eight slicers a day, right 
on target. But a shipment of cutting disks from a supplier was late 
and there was no inventory at the plant. Two members of the group 
visited the supplier (going around their own colleagues in our 
purchasing department) and discussed the problem with some of the 
supplier’s factory workers (who went around their colleagues in sales). 
Two days before the end of the month, the disks appeared. By then, 
however, they had run out of motors, which came from a plant 300 
miles away. Promises for delivery were made daily and broken just as 
frequently. Team members pestered the supplier and finally their 

Maverick a 89 

complaints were heard by a sympathetic executive. The motors 
arrived on the last day of the month. 

There were only a few hours left—not nearly enough to install all 
the motors. But the sheer team made a collective decision: each 
member clocked out at the end ofethe day, then returned to the 
assembly area—on his own time—and worked until 4:50 a.m., when 
the team had completed 210 slicers. Not one more, not one’less. 



For all the board room and locker room debate, most of the 
changes we had made so far were symbolic. But by late in 1985, the 
anxiety level had risen perilously among Semco’s middle managers. 
Fernando was gone, but his tough-guy philosophy was alive if 

“You’ve taken away my power,” supervisors would complain when 
they bumped into me in halls. I can t even tell if my people are 
arriving on time.” Or: “How do you expect me to meet this month’s 
sales goal when 1 don’t have control anymore?” 

They would tell me they were confused. They wanted me to 
explain Semco s policy. I kept thinking that was what was wrong. 
There was too much policy at Semco, and not enough thought, 
judgment, and common sense. But I understood why our managers 
were scared. 

If there was a group that was critical to our emerging plan, it was 
these midlevel supervisors. I never tired of telling them that they, 
more than I or my colleagues in top management, could be the 
biggest advocates—or the biggest obstacles—to our efforts to turn 
Semco into a more democratic and freer company. 

Modern corporations are typically shaped like pyramids, with the 
highest ranking executives at the top and progressively bigger groups 
of managers as you work down the hierarchy. Some pyramids are 
steep, like a bottle of Bordeaux. Others, the more efficient ones, are 

92 A Ricardo Semler 

flatter, like a bottom-heavy bottle of Portuguese rose. Either way, 
though, there is always a group of supervisors, department heads, 
and other professionals in the middle, no longer workers but not yet 
owners or shareholders. It isn’t unusual for these middle managers to 
be more zealous with authority (tight-fisted with the company’s 
money) than those at the top. 

We realized our middle managers needed to be stroked if we were 
going to succeed at what I was beginning to think of as a grand 
experiment. Like the audience in a Roman amphitheater, their 
thumbs up would insure that our efforts to increase the participation 
of Semco’s workers would thrive. Thumbs down and the lions would 
devour us. 

Our middle managers had studied at schools that taught tradi¬ 
tional organizational discipline and the importance of structure and 
supervision. They had been reared on competition and trained to 
accumulate symbols of power, such as parking spaces near the door 
and embossed business cards. They enjoyed saying things like, “See 
that four-door out there, the one with the graphite paint job? That’s 
my company car.’’ But democracy cannot be reconciled with expen¬ 
sive and unnecessary icons, especially those that come with radial 

As if the loss of perks and power wasn’t bad enough, our 
supervisors were further troubled by Semco’s growth. The acquisi¬ 
tions we had made meant that dozens of new managers and hundreds 
of new workers had to be incorporated into our organization. These 
people didn’t do things our old way, or even our new way. 

I decided to hold one large meeting at which our middle managers 
could air their concerns. It started at four o’clock one Friday 
afternoon. About forty managers were there. We started by asking 
them to state the issues they wanted to talk about and before we 
knew it had a list of forty items. Many concerned the authority 
they wanted back—being able to punish workers for tardiness and 
absenteeism, for example, or to control raises and bonuses for 

Tou have to decide what you want more, said a production 
manager at the food service equipment unit. “If it’s billing and sales, 
then you have to let us control our people. They’re beginning to 

Maverick a 93 

think they don’t have to obey us, because they can come and go as 
they like and they know we won’t fire them.” 

‘‘Your power to fire someone just because he’s late too often really 
determines your month’s invoicing?” Clovis asked. 

It sure as hell does, he shot back. “The problem with you, 
Clovis, is that you haven’t had to live down there with those guys. 

Give them a finger and they wont just take your arm, they’ll take 
your whole body!” 

Managers around the room were nodding. That’s what’s wrong 
with bosses, I thought to myself. So many of them are better 
prepared to find error and to criticize than to add to the effort. To be 
the boss is what counts to most bosses. They confuse authority with 
authoritarianism. They don’t trust their subordinates. 

You guys are unable to meet the budget you yourself determined, 
and now you’re using this as an excuse,” said Paulo Pereira, a tall, 
lanky young man who had come to Semco with Clovis and worked 
with him as our personnel manager. Unlike his soft-spoken mentor, 
he had a reputation for speaking his mind, no matter what. 

Bullshit! cried a hard-line production planning manager: “All 
this touchie-feelie stuff is bullshit.” 

Y?ah, cried a sales manager. It all comes from human resources, 
which spends the money that we work so hard to make.” 

On and on it went, with much gnashing of teeth and muttering 
under breaths. The managers wanted to know the rules of the game. 
We kept saying, “Well, let’s discuss it. Let’s just come to some 
agreement among ourselves.’’ But I could see it was only making 
them more anxious. 

Even though we had done away with the dress code, almost 
everyone at the meeting wore a suit and tie. (There were no women 
present.) A conservative bunch, they adhered to a rigid hierarchy, 
even when complaining. The first to speak were the most senior; 
then came those just below them, who would unfailingly support 
their superiors. 

1 admit I was more than a little to blame for this yes-man 
syndrome. Semco had been known to drive its managers hard and 
was quick to fire those who did not reach their goals. I myself had 
dismissed the entire tier of senior managers in one afternoon—no one 

94 A Ricardo Semler 

had forgotten that. So who could blame them for feeling angry, 
confused, and apprehensive now. 

I believed perhaps 20 percent of the middle managers were 
actually sympathetic to my efforts to make our factories and offices 
more democratic. I suspected another 20 percent laughed heartily at 
it, considering me a rebel with an inheritance. It was the rest of 
them who would determine whether I succeeded. 

We decided to hold another meeting for the middle managers the 
next week. And another one the week after. Then it became a weekly 
affair. The agenda was simple. We would talk only about company 
policy and philosophy. Operational problems were out. The hard¬ 
liners, who thought these sessions were useless, quickly dubbed them 
“Growing Bees in the Sky.” But I liked the term and used it officially 
from then on. 

As one session led to another, it became clear that we had two 
choices. We could write a completely new manual incorporating all 
our new policies and spelling out the emerging philosophy behind 
them. Or we could do away with the old manual and just not replace 
it. That way we would force people to make decisions based on 
common sense. 

Guess what we did? 


We tried to write new rules. We really did. But at every turn we 
found ourselves wading into a swamp of minutiae. Take our car fleet, 
for example. To reduce costs, we set up a central garage to make 
repairs, track mileage, and send people a memo when it was time to 
change the tires or renew the registration. But soon this new 
department was ensnared in conflict. Peoples cars would break down 
and they would call the garage to complain. Or someone urgently in 
need of his car would arrive at the garage to find there weren’t any 
vehicles available. Department heads who received bills for service on 
their cars, discovered they hadn’t budgeted enough, and then spread 
the word that the auto department was inefficient, or maybe even 
taking kickbacks from greedy local mechanics who were making the 

From this we quickly concluded that some departments were 

Maverick a 95 

better not created and some rules were better not written. Common 
sense would be the best alternative, by far. But we were careful. We 
didn t want to rattle an already jittery crew. There was no grand 
announcement of our decision. Over the next three or four months 
we simply collected all our procedures manuals. (Many months later 
we did the same with our organizational chart, not that we had 
consulted it much.) People would ask us from time to time when the 
new manuals would be ready. Eventually, some began to suspect that 
an update wasn’t going to appear and asked us why. Only then did 
we say aloud what we had been thinking: that we were trading 
written rules for common sense. 

And that is the system we have today, which is barely a system at 
all. When you get a company car at Semco, you can do anything you 
want with it. If you have a friend who is a mechanic, have him take 
care of it. We want our employees to treat Semco’s vehicles as they 
would the family sedan. We’re comfortable having their judgment 
apply to our car as well as theirs. 

But this was all terribly frightening to some of our people, 
especially at first. We had to do lots of hand holding. Clerks in our 
finance department, for example, weren’t comfortable deciding how 
much Semco should keep as minimum balances in its checking 
accounts. ‘Just think of them as your own accounts,” I would 

‘‘But Semco has so many accounts,” they would reply. ‘‘What 
happens if we have an emergency and need cash?” 

And I would take a deep breath, and implore them, one more 
time, to use their common sense. 

Where did all these rules come from, anyway? 

They were, I suppose, an unhappy by-product of corporate expan¬ 
sion. How does an industrial giant act as it grows? First, manage¬ 
ment concludes that a company cannot depend on individuals. After 
all, they have personalities and finite life spans. A corporation is 
supposed to be impersonal and eternal. 

Next thing you know, committees and task forces and working 
groups are spewing out procedures and regulations and stomping out 
individuality and spontaneity. 

96 A Ricardo Semler 

In their quest for law, order, stability, and predictability, corpora¬ 
tions make rules for every conceivable contingency. Policy manuals 
are created with the idea that, if a company puts everything in 
writing, it will be more rational and objective. Standardizing meth¬ 
ods and conduct will guide new employees and insure that the entire 
company has a single, cohesive image. And so it became accepted 
that large organizations could not function without hundreds or 
thousands or tens of thousands of rules. 

Sounds sensible, right? And it works fine for an army or a prison 
system. But not, I believe, for a business. And certainly not for a 
business that wants people to think, innovate, and act as human 
beings whenever possible. All those rules cause employees to forget 
that a company needs to be creative and adaptive to survive. Rules 
slow it down. 

Semco had a particularly complicated set of rules on travel 
expenses. Our auditors often spent hours arguing over whether 
someone on a business trip should be reimbursed for movie tickets. 
Well then, what about theater tickets? What would we do if an 
employee went to a concert that cost $45? Or $ 100? And what about 
calling home? How often should the company pay? Was a five- 
minute call reasonable? Vfhat it an employee had, say, four children? 
Are seventy-five seconds per child sufficient? 

See what I mean? 

Without rules all answers are suggested by common sense. No, I 
can’t define what common sense is, but I know it when I hear it. 
Some of our people stay in four-star hotels and others, sometimes 
with much higher salaries, choose lesser digs. Some people spend 
$200 a day on meals; others get by on half as much. The point is, if 
we can’t trust a manager to use good judgment about such things, 

we sure as hell shouldn t be sending him off to do business in our 

A company makes, sells, bills, and, God willing, collects. It 
doesn’t need to know if the taxi ride being claimed by a manager was 
for business. Or if another manager couldn’t have stayed in a hotel 
with three rather than four stars. With few exceptions, rules and 
regulations only serve to: 

Maverick a 97 

1. Divert attention from a company’s objectives. 

2. Provide a false sense of security for executives. 

3 . Create work for bean counters. 

4. Teach men to stone dinosaurs and start fires with sticks. 

The desire for rules and the need for innovation are, I believe, 
incompatible. (Remember, Order or Progress.) Rules freeze com¬ 
panies inside a glacier; innovation lets them ride sleighs over 

Weve found we can replace nearly every rule that those master 
sergeants called comptrollers can impose. This does not mean that all 
written instructions are forbidden—but our people are not afraid to 
ignore procedures that don’t seem applicable or wise. There are no 
absolute truths at Semco, nor are we out to make everyone do things 
the same way. 

There is another, less obvious dividend to the banishing of rule 
books: people begin to make more decisions on their own, decisions 
they are usually better qualified to make than their supervisors. 

What about the economies of scale? I am often asked. If Semco 
went out and bought tires for all the cars in our fleet at once, for 
instance, wouldn t we get a better price than if employees go out and 
buy tires on their own? Probably. Then again, if our auto department 
scrupulously and blindly followed the manufacturer’s overly conserva¬ 
tive recommendations on tire replacement, we would no doubt buy a 
lot more tires than we do now by letting our employees decide when 
they need changing. And I’ll be that’s true with many items we used 
to buy corporately and now purchase individually. 

Of course, I don’t know for sure, since under our no longer 
neurotically compulsive budget system we don’t keep track of these 
purchases anymore. But if you want my advice, take a deep breath, 
pluck up your courage, and feed the policy manual to the shredder, 
one page at a time. Let companies be ruled by wisdom that varies 
from factory to factory and worker to worker. To do otherwise only 
gives those tough-guy comptrollers the comfortable feeling that the 
company is organized, and provides jobs for dozens of disturbed souls 
who should be retrained for some useful purpose. 

At first it was hard for us. But with a great deal of commiseration 

98 A Ricardo Semler 

and consultation the shock of rulelessness began to subside, and our 
middle managers began to remove their armor plates. 

I liked to tell them that a turtle may live for hundreds of years 
because it is well protected by its shell, but it only moves forward 
when it sticks out its head. 



They had been there for hours, through the lazy, steamy, Brazilian 
afternoon. On one side of the long wooden cafeteria table were two 
union directors and three shop-floor workers. Across from them were 
three Semco executives. They were all trying to settle a week-long 
strike that, like most labor disputes, revolved around money. But as 
befits a company where traditional corporate dogma was being 
discarded and unpredictability was a way of life, this walkout began 
when Semco voluntarily launched a program to give our workers a 
raise they had neither asked for nor were expecting. 

In the three years we had owned the Hobart plant we had proudly 
watched as workers assumed more of a role in determining how it 
was run. They set their own production quotas and redesigned the 
slicers, peelers, and dough mixers they made. Now tension hung 
over the place like a vulture awaiting the lions last mouthful. The 
long silences at the bargaining table indicated the sides were far 

The strike was our eighth in seven years, but all had been brief, so 
we were hardly experts in resolving them. Clovis was our chief 
representative at the bargaining table, and his thirty years of mana¬ 
gerial experience told him what was on the minds of the men across 
the table. Workers had practically been bred to distrust anyone who 
represented money—even someone like Clovis, who had built a 
career trying to dissipate such feelings. We had come a long way at 

100 A Ricardo Semler 

Ipiranga, but in times of stress those early, bitter feelings just seem 
to reassert themselves. Clovis leaned back, smoothed his mustache, 
and looked across the table. What would it take, he thought to 
himself, to create trust if all that had been accomplished here had 
failed to do it? How ironic that the walkout was sparked by our 
efforts to insure that the wages we paid were fair. 

Clovis s chief adversary was the factory committee leader, Joao 
Soares. Although aware of his charisma, Joao had thought of shop- 
floor representatives as either embittered opponents of capitalism or 
impassioned leaders. He didn’t see himself as either of those. He had 
suffered through a series of hard-edged factory jobs, but after a few 
years at Semco had begun to think he had, at last, found a company 
that would let him have his dignity and even his aspirations. Now, 
sitting across from Clovis, Joao wasn’t sure. 

Given the fits and starts of Brazils hyperinflationary economy, we 
were no longer sure our pay scale was above average, which was 
where we wanted it if we were to keep our work force stable. So we 
embarked on a program to reevaluate the wages we paid at all levels 
of Semco. Taking it one business unit at a time, we thought we 
could adjust the entire company in two years. 

Most Brazilian companies determined wages and salaries by refer¬ 
ring to orthodox salary models developed in the United States and 
Europe. At best, these impersonal sets of statistics were applied by 
pseudo-experts who, if they even visited a company, rarely listened to 
its employees. The most sophisticated of these plans had nowhere 
near the complexity needed to gauge our participative work force or 
peculiar corporate culture. But we didn't like them for a more basic 
reason: they were one-sided. We wanted our people to be involved in 
the salary-adjustment process. It was their paycheck, after all 
So we turned to Paulo Pereira, our resident remuneration expert, 
who, as we might have expected, came to a radical conclusion. The 
only truly correct pay, he asserted, was an average of what a worker 
thought he should receive and what a company could afford to pay 
That was for business schools, not factories, he realized. But Paulo 
an unrepentant idealist, set out to find this point of balance anyway.’ 

In true Semco fashion, his first step was to ask about thirty factory 

Maverick a 101 

and office employees to visit other manufacturing companies that 
were about the same size as Semco. They were to interview their 
counterparts there, comparing skills, responsibilities, and paychecks. 
It sounds so utterly sensible to ask workers to do this, but it just 
isn t done. Indeed, at first we had trouble convincing some corpora¬ 
tions to receive our workers. But eventually our people collected 
comparative information on every job at Semco, from general man¬ 
ager to janitor. And sure enough, we had fallen behind, no doubt 
because we had so little turnover. 

The marine equipment plant at Santo Amaro was first to be 
adjusted, because we believed it lagged the most. Right after would 
come the food service equipment unit. 

At the end of each month there was a barbecue at Ipiranga for all 
the employees. As it happened, a worker from Santo Amaro had just 
been transferred to the food service equipment unit and spent his 
first afternoon at the barbecue, where he proudly told his new 
co-workers he had just received a 40 percent raise. Almost everyone 
at Santo Amaro got one, he said. 

As he made his way around the gathering, indignation rose. 
When the party ended, small groups of irate workers stayed 
behind to discuss the outrage. How could workers at Santo 
Amaro have received a 40 percent raise and those at Ipiranga get 

How indeed? 

The next day, a Saturday, was the end of the month. Dozens of 
workers were at Ipiranga to meet production quotas. It was an 
understatement to say their usual enthusiasm was dampened by Santo 
Amaro s good fortune. Since the workers didn’t require supervision, 
no one from management was on hand to explain the full story of the 
salary adjustment. 

After listening to the grumbling, Soares tacked a note to the food 
service equipment unit’s main bulletin board. “We have been tricked,’’ 
it said. “Santo Amaro has received a 40 percent raise and we didn’t. We 
must go on strike at once, until we are fairly treated.’’ 

When the managers arrived Monday morning, the strike was on. 

It didn t matter that the facts had been garbled—that the average 

102 A Ricardo Semler 

increase at Santo Amaro had been 8 percent, not 40 percent, and that 
any raise above 15 percent was to be phased in. Nor did it matter 
that the food service equipment unit was next in line for readjustment. 

Soares talked to the regional union director, Geraldo Mello, who 
was at the plant that morning, and he was even more outraged than 
Joao. “They can’t favor their other workers and ignore the Hobart 
people, Mello told him. “We must teach them a lesson.’’ 

When some executives at the plant heard that, they had the classic 
response: they wanted to make the strikers’ lives as miserable as 
possible. They suggested that we hold on to the paychecks for the 
period before the walkout and lock the gates so the workers couldn’t 
sit around the plant, as they all were doing. They even wanted us to 
suspend health care benefits. But cooler heads thought we should 
maintain the Semco way of doing things, no matter how we were 
provoked, and to my relief they prevailed. 

I had come to believe labor unions were more than a necessary 
evil. They are one of the few legitimate agents of workplace change. 
Not all union leaders are sensible, nor is every union’s position 
reasonable. But to pretend a union doesn’t exist, or to try to defeat it 
whenever possible, with whatever means, at whatever cost, is hardly 
worthy of the term “strategy.” Executives in the United States who 
take comfort in the sharp increase of nonunion shops are myopic. It’s 
wise to keep all the antennas connected. The ostrich that buries its 

head in the sand has a bigger problem than limited vision; its rear 
end is an enormous target. 

We all know the usual ways of combating a strike: 

1. Take a stand. Show the flag. Don’t back down. 

2. Guarantee that anyone who ivants to ivork can, even if that 
means calling in the police. 

3 . Protect company property, with force if necessary. 

4. Make it hard for the workers by closing the plant and 
suspending benefits. 

3. Try to divide and conquer the strikers. 

6. After it's over, fire the instigators and anyone else you want to 
get rid of, intimidating others in the process. 

Maverick a 103 

These recommendations are shortsighted and costly. Troublemak¬ 
ers, like weeds, will always sprout up, because management does 
such a thorough job putting down fertilizer. When we acquire a 
company or start up a new factory, one of our first moves is to invite 
the local union leaders over for a visit. Hobart had thanked the 
heavens it wasn’t discovered by organized labor, but we asked the 
Metalworkers Union, which has earned its reputation for com¬ 
bativeness, to organize the plant and be part of the factory com¬ 

Recognizing the existence of unions does not automatically mean 
agreeing with them. We have courteously received union committees 
just to say no to eighteen of their nineteen demands. But we 
recognize the renewing power of unionism, and the importance of 
not becoming ostriches. He who seeks to keep the union outside the 
factory gate will soon need pants with reinforced seats. 

During a strike, we follow these rules: 

1. Treat everyone like adults. 

2. Tell the strikers that no one will be punished when they return 
to work. Then don't punish anyone. 

3 . Don't keep records of who came to work and who led the 

4. Never call the police or try to break up a picket line. 

3. Maintain all benefits. 

6. Don t block workers' access to the factory, or the access of union 
representatives to the workers. But insist that union leaders 
respect the decision of those who want to work, just as the 
company respects the decision of those who don't. 

7. Don't fire anyone during or after the strike, but make everyone 
see that a walkout is an act of aggression. 

No, I m not a graduate of the Patrice Lumumba Institute in 
Moscow, and to prove it let me say that at the time of the Ipiranga 
strike Semco only accepted an unconditional return to work. Al¬ 
though we valued dialogue as no other company did, we never 
negotiated during a strike. 

This was our thinking: almost all companies that follow the first 

104 A Ricardo Semler 

set of rules end up bargaining under pressure and making concessions 
they later regret. In other words, they start out like Arnold 
Schwarzenegger and end up as Woody Allen. Under our procedure, 
there was only one end to a strike—the return to work by everyone, 
with no pay for hours not worked and no concessions made during 
the stoppage. Only after the workers come back would we resume 

So we allowed the striking employees to enter the plant and use 
the cafeteria for meetings. Breakfast and lunch were served as usual, 
subsidized by us as usual. Medical and other benefits continued. But 
we wouldn’t negotiate. 

As the strike dragged on, executives would have informal conver¬ 
sations with workers in the cafeteria, trying to win them over. 
Anyone who ventured into the plant would have thought it had 
merely run out of raw materials. People were sitting around playing 
cards and talking. There was no sign of animosity. Some workers 
wanted to return to their jobs, believing us when we told them we 
intended to raise their salaries soon. Most didn’t. 

Then Mello, the regional union director, decided it was time to 
turn up the heat. Without consulting Soares or the other factory 
committee members, he went to court and sued us for discrimina¬ 
tion. Pie was convinced a judge would compel Semco to agree to 

some kind of an increase, giving the workers and the union a 

By the time everyone heard about it, the case was on the calendar. 
Soares felt humiliated. The strike was being handled professionally 
and even courteously, and going to court was regarded as much too 
extreme a tactic. Moreover, anything can happen in a courtroom. As 
we say in Brazil, from the behind of a baby and the head of a judge, 
no one can know what’s coming. 

"Let s make a deal, Clovis," said Mello, when the union leader 
returned to the plant from court. 

“The company will comply with any decision reached by the 
judge,” was Clovis's clipped reply. His experience, along with 
intelligence from the shop floor, suggested that many workers were 
now willing to accept a 5 percent increase, down from the 40 percent 

Maverick a 105 

they had been demanding. There was a risk that the judge would 
order a larger raise than 5 percent, but we would take our chances. 

So everyone headed for court. After deliberating for three hours, 
the judge ruled that the strike was illegal. But if the union had lost, 
we hadn’t yet won, since the workers still weren’t working. 

The next day the mood at the plant was funereal. By late afternoon 
several workers sought out Joao and asked him to talk to manage¬ 
ment about getting paid for the unworked days. We turned that 
down, but offered the strikers a chance to work overtime if they came 
back, so their next paycheck wouldn’t be smaller than usual. They 
accepted, and the strike was over. 

At week’s end, Clovis suggested that the managers and strike 
leaders sit down together to analyze the walkout. This session, which 
lasted four hours, became known as the meeting at which “The 
Bananas Ate the Monkeys.’’ We said we were sorry for our inflexibil¬ 
ity. To prove it, we revoked our policy of not negotiating during 
strikes. Joao and his people confessed that the note on the bulletin 
board had been a mistake because it left the workers no choice but to 
strike. Had a stranger entered the room, he would have thought that 
the managers were dressed in overalls and the workers in suits and 

The strike taught both sides large lessons. We realized that being 
participative was not enough. We would have to learn to communi¬ 
cate better, because as much as anything people’s perceptions gener¬ 
ate strikes. The workers realized that walkouts were hardly an 
effective method to solve problems, and they have become rare at 
Semco since then. 

A few months after the walkout, incidentally, the new salary 
schedule was implemented at the food service equipment unit. The 
average increase was 18 percent. No wonder the workers were upset. 


Almost all businessmen think their employees ere involved in the firm 
and are its greatest asset. 

Almost all employees think they are given too little attention and 
respect, and cannot say what they really think. 

How is it possible to reconcile these two positions? 

The sad truth is employees of modern corporations have little reason to 
feel satisfied, much less fulfilled. Companies do not have the time or the 
interest to listen to them, and lack the resources or the inclination to 
train them for advancement. These companies make a series of demands, 
for which they compensate employees with salaries that are often consid¬ 
ered inadequate. Moreover, companies tend to be implacable in dismissing 
workers when they start to age or go through a temporary drop in 
performance, and send people into retirement earlier than they want, 
leaving them with the feeling they could have contributed much more had 
someone just asked. 

The era of using people as production tools is coming to an end. 
Participation is infinitely more complex to practice than conventional 
corporate unilateralism, just as democracy is much more cumbersome 
than dictatorship. But there will be few companies that can afford to 
ignore either of them. 




“A great business,” Henry Ford once said, “is really too big to be 

Makes you wonder, doesn’t it? Old Henry was proud of Ford 
factories like River Rouge, with an assembly line that seemed to 
stretch forever. Today these behemoths are endangered, if not obso¬ 
lete. Gigantic enterprises are “downsizing” as fast as they can. They 
have found that the rationale that encouraged their growth, the 
economies ol scale, does not hold forever. Get too big and you 
quickly discover the diseconomies of scale. 

Large, centralized organizations foster alienation like stagnant 
ponds breed algae. In massive corporations, an employee will know 
lew of his colleagues. Everyone is part of a gigantic, impersonal 
machine, and it is impossible to feel motivated when you feel you are 
just another cog. Human nature demands recognition. Without it, 
people lose their sense of purpose and become dissatisfied, restless' 
and unproductive. Stalin understood this. Prisoners in his gulags 

were obliged to dig enormous holes in the snow, then fill them in. It 
broke their spirits. 

As Semco grew, we had begun to suffer some symptoms of 
gigantism, though we didn t recognize them at the time. Our 
salespeople, for example, insisted on filling out complex forms each 
time an order was received, stipulating exactly what the customer 
wanted, rather than just explaining the order and its peculiarities to 

110 a Ricardo Semler 

our engineering, purchasing, and production people. There was 
nothing inherently wrong with this, except that it generated an 
enormous amount of paperwork. But the salespeople felt their 
department would be better protected if they could prove, in 
writing, that a customer didn’t receive exactly what he wanted. Then 
our engineering department decided it would only hear complaints 
from shop-floor workers on Tuesdays and Thursdays from 2 p.m. to 4 
p.m., rather than simply talking to workers whenever they came 
around. The teamwork that was flourishing at the food service 
equipment plant wasn’t catching on elsewhere in the company. In 
fact, cliques were forming among managers, and much of Semco 
seemed to act distressingly like a government agency. We were even 
keeping visitors waiting at plant gates and reception areas. 

We reacted by making the moves the classic organizational studies 

Semco s structure was what business school professors call a 
functional system. That meant production managers at our plants 
reported to the production director at our headquarters; the salespeo¬ 
ple answered to the marketing director; the administrative officers to 
the financial director; and so on. It sounds orderly, but anyone who 
has worked in a diversified, multiplant company knows that a high 
percentage of decisions made under this long-distance arrangement 
are just plain wrong, and take too long as well. It is a feudal system, 
isolating engineering from sales and sales from finance and generat¬ 
ing solutions and strategies that serve one department at the expense 
of another. 

No doubt because of this, European companies seem to prefer an 
organization based on a matrix system. Recognizing that a computer 
programmer at a distant plant can t possibly be instructed effectively 
solely from the administrative director’s office, the lucky chap is put 
under dual command, reporting to his local plant manager as well. 
The matrix system provides a manager with two sources of informa¬ 
tion, two types of expertise, two perspectives on a particular problem 
or goal. It allows for a complete picture. But what if the head office 
wants something done that the plant manager feels isn’t necessary? 
Multiply this conflict by hundreds or thousands of employees and you 

Maverick a 111 

can see why people in matrix organizations are skilled diplomats, but 
not necessarily great businessmen. They have learned to survive in an 
environment that resembles the United Nations by behaving with 
extreme caution at all times. Of course, you can’t try new ideas 
without taking risks or making mistakes, and people in companies 
organized under the matrix system generally don’t. 

Alas, I know all this from experience. So in 1986, after a flirtation 
with the matrix system, Semco tried yet another organizational plan: 
autonomous business units. We split the company into separate 
units, with a general manager for each. Theoretically, all the units 
would be independent. That was bound to breed competition be¬ 
tween them, but we believed it would be a healthy competition, 
unlike interdepartmental warfare. 

As a practical matter, however, this autonomous system led us 
back to a functional organization. That was because the autonomous 
business units needed some form of internal coordination and they 
chose what might be called a miniaturized functional system, in 
which managers inside each unit reported to the general manager. 

And guess what happened next? In no time at all, our managers 
were studying matrix systems, to alleviate the tension. 

Around and around we went, and where we would stop nobody 

But wait. It was the dawn of the computer era. Maybe microchip 
management would save us. 

Semco first flirted with computers back in 1980, when Harro and 
I went to New York and ordered an IBM PC. We had to wait three 
months for delivery, then learn commands in BASIC to run it. Even 

so, we soon had concluded that we had no choice but to computerize, 
and quickly. 

Trouble was, we were convinced by our computer specialists that 
we needed to buy powerful machines that could talk to each 
other and hire highly paid, difficult to get, and harder to keep 
information managers to run them. We even entertained the idea of 
hiring a systems vice president to coordinate these information 
gurus a prince of processors. As a matter of fact, we ran after one 
candidate, desperately trying to pull him away from a French 

112 A Ricardo Semler 

company with all kinds of promises. By the time we were finished 
negotiating, he had a salary darn close to mine and someone to sweep 
the floor in front of him as he walked around. Thank God he didn’t 
accept our offer. We did buy, lease, rent, and steal almost everything 
the salespeople from IBM, Bull, Fujitsu, and Hewlett-Packard 
showed us. But hey, we were doing all the trendy things then. We 
even had a Japanese strategic planning manager. 

A master plan hundreds of pages long was drawn up listing the 
hardware and software we needed. No doubt about it, our informa¬ 
tion systems department was going to be the brains of the company. 
We hired analysts, digitation clerks, and programmers. We bought 
the smartest workstations we could find and hooked them to central 
data-processing servers that would link all our plants. People who 
used pencils to update inventory cards as new supplies arrived would 
now punch codes into computer terminals. Or would they? To make 
sure, we started brainwashing courses to turn our clerks into twenty- 
first-century apparatchiks. 

In no time our old systems were no longer operational. Too bad 
our new, electronic systems weren’t operational, either. We waited a 
month, two months, a year. There were always excuses—the training 
program wasn’t finished, the hardware had glitches, the software had 

bugs. It took us too long to realize we were heading down a yellow 
brick road. 

I remember a tour I took of the information system at the Ipiranga 
plant when I was working there. 

"Were close to being on the air," the resident computer jockey, 
Wilmar Fagundes, told me proudly. "You know, we just got rein¬ 
forcements last month. We needed four programmers, three systems 
analysts, and two digitation clerks, but we were short one of them. 
To run MRPS." He must have noticed my blank expression. 
"Manufacturing Resources and Planning System” 

I just nodded. 

"So now were hooked up to an IBM 3090, wrth twelve terminals 

to be installed, Wilmar went on. “Here are the production planning 

I gave him another blank look. 

“See, here," he said, spreading out a pile of enormous white sheets 

Maverick a 113 

of paper covered with hundreds of tiny boxes. Wilmar was still 
working on the system, so the design had to be done manually before 
it could be fed into the machine. "Every item in production is 
broken into subsets. Here, this is a simple one: a dishwasher. We 
have fifteen sizes of dishwashers, with eleven variations of each size. 
Each has approximately 300 components.” 

“And that’s a simple one?” I said. 

Oh, yes. At the Santo Amaro plant, where well be installing the 
system next, there are products such as the biscuit machinery line. It 

has 12,000 components, eighty variations of twenty-five different 

Anyway, lets take an AM-12 dishwasher. Here it is”—Wilmar 
pointed—"this list includes all of its subsets. Like branches of a 
tree, right? Each branch has its own branches, the components. And 
the components branch out into twigs. 

Here, this is a dishwasher arm. You know, the hollow metal tube 
that sprays the hot water onto the dishes. We have that tube in 
various kinds of metal. See, this branch is the stainless steel tube.” I 
watched his finger slide along. "This stainless steel arm has a code. It 
will tell us if this arm is made here or bought from a supplier or if 
some of it is made here and some of it is bought from a supplier. 
This arm is in the third category, incidentally. So let’s look at the 
production process it goes through, right here on sheet number 267 
Urn, 256, 258. Where is 267?” 

Wilmar muttered under his breath as the orderly stack of white 
sheets began to fly around the table. 

I spent three and a half hours with Wilmar. I saw how the stainless 
steel arm was ordered by computer, how raw material was set 
aside if available or reordered. How the ordering procedure re¬ 
quired three different quotes. How the winning bid was decided by 
weighing statistical data on delivery, price point ratios, and payment 
terms. How inventory items were divided into categories by weight, 
cost, and the space they took up in a storeroom. How the nuts and 
washers that held the arm together were made from steel plates that 
were stocked in the yard and then moved by the computer through 
seven different areas of the factory and three production procedures— 
cutting, hole boring, and welding. 

114 A Ricardo Semler 

What a hard way to make money! 

I’m sure the people who devised the system envisioned an auto¬ 
mated, machine-managed plant where steel plates were gracefully 
eased through the rear door and elegantly crated dishwashers rolled 
out the front. Except it never happens that way. I’ve never been to a 
plant that doesn’t have too many nuts for too few bolts, or shelves of 
some part for a product it no longer makes. Even in plants that 
function tolerably well, workers become servants to the Production 
Dragon, which is fed tons of parts and spits out finished products. 
This Dragon only smiles when people don’t mess with it, or, for 
God’s sake, try a new way of doing things. Sounds like Henry Ford 
in 1908, doesn’t it? 

When computers first arrived, they were mostly used for tasks that 
required the manipulation of huge amounts of information, such as 
conducting a national census, keeping track of military movements, 
or untangling huge social service programs. Now they have evolved 
to the point where a machine that used to take up a whole room 
occupies a lap, or even a pocket, and they have insinuated themselves 
into just about every facet of business. People can’t think without a 
keyboard at their fingers and a monitor in front of their nose. But 
something curious has happened: computer-generated information, 
which was the means, has become the end. Instead of helping us 
organize data, computers are drowning us in it. 

The explosion of computer-generated information began at a time 
when the companies were run by executives who weren’t part of the 
computer age, and so could be wrapped around the fingers of 
computer professionals, who leveraged their special knowledge into a 
sort of priesthood, inventing enchanted words and sacred hymns, 
throwing hardware and software at people more comfortable with 
Tupperware, and making most of us feel like dimwits. Older 
executives, especially frightened, thought to themselves: “I’ll hold 
back this stuff as long as I can, and when that’s no longer possible I’ll 

pretend that I understand what they say and buy as much of it as I 

The result has been a grossly inappropriate use of these extraordi¬ 
nary machines. Reports are fattened up with charts and graphs and 
thousands of numbers, most of them unneeded and unheeded. 

Maverick a 115 

Everything has become excessively complicated and confused, which 
is exactly the condition computers were supposed to remedy. I can 
shamelessly say that Semco was swept along on the electronic tide, 
going from the Stone Age to the Space Age in just a few years. But 

the more I learned about microchip management, the more I came to 
fear it. 

I once visited a small unit of ours three days before the end of the 

“What do you think this month's billings will be?” I innocently 
asked the resident computer wonk. 

We already know. We have the number here, in the terminal.” 
How can you.' There are still three days left in the month.” 
‘Oh, no. We stop billing four working days before the end of the 


I was shocked. It meant inventory was tied up longer than it had 
to be, increasing interest costs, and orders would sit for days before 
leaving the plant, increasing customer frustration. Just picture 

Federal Express stopping for a few days each month to feed paperwork 
to its computers. 

Well, we have to fill out the basic invoice, send it to the head 
office, wait for batch processing on the central computer—” 

Wait, I interrupted, why can t you just issue the invoices on 
your own terminal?” 

Because then they wouldn’t be automatically recorded in the 
accounts receivable program, which cross-references all the invoices.” 
“I see,” I said, although I didn’t. “Go on.” 

The lot is batch processed after all invoices from all our units are 
gathered—we run them in numerical order, for tax purposes—and 

then our invoices are returned to us. And then we can ship the 

“How did you do all this in the past?” 

Oh, that was very primitive. We would wait until the last 
minute, then type out the invoice. Sometimes we would be here in 
the middle of the night, getting invoicing out of the way, to make 
the month’s sales larger.” 

“How many invoices did you issue then?” 

“About 150.” 

116 A Ricardo Semler 

“And now?’’ 

“About 120.’’ 

Two days later the unit was off the computer and back on the 
primitive, manual system. And soon invoicing rose 15 percent, as 
employees got back to making last-minute shipments in all-out 
efforts to move finished products out the door. And within a month 
all the other computer terminals at our business units had been 
returned to headquarters, and our mainframe there was disconnected. 
We no longer have all those programmers or keypunch operators; we 
have dismantled our information systems department and thrown out 
our systems master plan. We gave all our techies an opportunity to 
make a living elsewhere and sighed as we sat back and relaxed. In 
typical Semco fashion, whoever decides he needs a computer goes out 
and buys one. Whatever anyone feels is necessary is all right with us. 
The catch is, everyone has to learn to operate it themselves. Our 
worries about making one computer compatible with another are 
over. It’s every microprocessor for itself and to hell with the econo¬ 
mies of scale. 

This much seemed clear: either you can adopt sophisticated, 
complex systems to try to manage the complications, or you can 
simplify everything. 

Finally, we chose the other path. 

And we had the perfect guide—Joao Vendramin, our balding, 
bespectacled, in-house economist and wise man, whom we had the 
extreme good fortune to inherit in the Hobart deal. Though he had 
occasional outbursts, Joao was usually reserved and thoughtful. He 
certainly looked relaxed, having abandoned suits and ties. I can still 
remember the snickering when Joao would sit at our mahogany 
conference table, his feet atop the polished surface. Ever careful about 
expensive furniture, he would always take off his shoes first. 

Joao would lean back in silence. Then he would suddenly break 
in, quoting Aristotle or Machiavelli or offering up his own offbeat 
pronouncement. It got so people would defuse the tension during our 
meetings by saying things like, “As Joao would say about our 
production schedules, you can’t fly bicycles in strato-cumulus clouds.’’ 
And he would just smile. 

Maverick a 117 

Who better to send around the world than Joao? Who better to 
visit companies that were breaking rules, or eliminating them, and 
setting out on their own? These were the companies we felt could 
help us solve our organizational problems. 

So off Joao went to Sweden, where he watched assembly line 
workers at Volvo who, instead of performing one task, worked in 
small teams to put an entire car together. (This system fell out of 
favor years later, but largely because of union and management 
problems, not because of a lack of teamwork.) In Delaware, he 
visited W L. Gore, a company that had discarded conventional 
organization charts and created what they call a ' lattice organiza¬ 
tion, with nonhierarchical relationships between jobs. Gore had 
stripped the titles from managers’ business cards and adopted a 
system in which salaries corresponded to monthly performance and 
were adjusted by groups of employees who had the option of cutting 
an individual’s pay to zero (which, according to the late Bill Gore, 
tended to discourage further work from the party involved). At 
Samsung in Korea and Toyota, Kyocera, Sharp, and TDK in Japan, 
Joao saw other ways in which modern production methods and 
worker involvement were joined, although their systems depended 

on such cultural traits as submissiveness and veneration of the 

‘There is no way to treat employees as responsible and honest 
adults unless you let them know and influence what is going on 
around them, Vendramin told us when he returned from his world 
tour. And there is no way to let them become involved in the 
decisions that affect them if the plant they work in has too many 

Yes, there are schemes and mechanisms to convince people that 
they matter, but they don t work for long. At some point the workers 
notice that they are never consulted about the really important 
decisions. The only way to change is to make each business unit 
small enough so that people can understand what is going on and 
contribute accordingly.” 

According to Vendramins diagnosis, Semco was already afflicted 
with an acute case ol bigness, brought on by our acquisitions and our 
success. The cure was logical, if unfashionable in an age of conglom- 

118 A Ricardo Semler 


eration. Factories that had become too large for their own good 
should be broken into units small enough to insure that the people 
who worked in them would feel human again. In a small factory, it is 
possible to know everyone by their first name, to debate plans and 
strategies, to feel involved. To belong. 

Vendramin’s proposal was that Semco should replicate itself like an 
amoeba. So we prepared to divide. 



It s airy lairy, sniffed Henrique Pinto, the dark-haired, mustached 
general manager of our Santo Amaro plant and one of a dozen or so 
managers gathered in a third-floor conference room. 

“Airy-fairy?” I asked. 

“The advantages you’re listing on the board can’t be measured, but 
the disadvantages are concrete. And costly. I call that airy-fairy.” 

He s right, said Clovis. The cost of duplicating security guards, 
receptionists, secretaries, and all the rest is easy to add up. Motiva¬ 
tion and the feeling of belonging can’t be quantified.” He paused. 

But that doesn t mean that the monetary value attached to the easily 
measurable items is greater.” 

Then how will we know if we are making the right decision?” 
Henrique wondered. 

“We won’t,” I said. “It’s really a leap of faith.” 

“We’ve taken leaps of faith in the past, and fallen,” Henrique 
k^oke in. Like investing in Flakt. It s been three years since we 
bought it, and it will take us five more, at least, to see a profit.” 

You re right, Henrique,” said Vendramin, who was sitting with 
his shoeless feet on the conference table (as usual), staring out the 
window (as usual). But we made the same decision about the biscuit 
machinery line. We spend hundreds of thousands of dollars before 
seeing a return. Then we recovered everything in a single year.” 

120 A Ricardo Semlf.r 

"So, what now?" Henrique persisted. "Do we go ahead and divide 
all the plants, no matter what the cost? What is it you call it, Joao?" 

The amoeba approach," Vendramin said with a smile. 

We had been discussing the amoeba approach for weeks, with no 
discernible progress. The engineers and other technically oriented 
souls, like Pinto, were skeptical about the gains that might result 
even if by making our plants smaller our workers somehow became 
more productive. But our problems—the meetings that ran too long, 
the power struggles between groups of managers, the impossibility of 
making everyone in a plant feel like they were a part of something, 
the alienation I still saw ail around Semco, the lack of cathedral 
builders—weren’t going to go away by themselves. They were, I 
believed, rooted in bigness, the cure for which, if expensive, was 

simple. We would divide ourselves like an amoeba—and, we hoped, 

In some cases, it would mean moving workers to a new factory. 
Other plants would just be split down the middle, like the motel 
room Clark Gable and Claudette Colbert shared in It Happened One 
Night. In any event, the new units, like those from which they 
sprang, would be fully autonomous. Managers would be free to 
manage as they wished, Semco-style. Our centralized corporate staff 
would provide support in such areas as accounting and human 
resources and, when asked, offer opinions on strategy. But if a unit 
didn’t need the people at headquarters, that was fine with us. If a 
plant achieved the performance we all expected, months would go by 

before it might receive a visit from any of us. They would deserve 
our benign neglect. 

There were huge costs associated with Vendramin's amoeba plan 
as the other number crunchers kept reminding us. What was now 
one factory would have two of everything, from janitorial staffs to 
loading docks and from data-processing departments to parking 
areas. The two plants would also require that we carry more 
inventory, in two storerooms. And then there were big-ticket items 

such as the cost of additional machines and buying or leasing new 

Sometimes it seemed foolish even to me. But we went ahead 

Maverick a 121 

anyway, basing our decision on two feelings corporate managers are 
usually afraid to trust: intuition and faith. 

In 1986 Brazil underwent a drastic economic transformation. A 
new economic policy froze prices and cut inflation from 25 percent a 
month to zero. The finance minister, a gaunt, serious, quixotic man 
named Dilson Funaro, was intent on taming an inflationary culture 
in which everyone raised prices in anticipation of coming rises, 
perpetuating an unending cycle of increases. Funaro was terminally 
ill with lymphatic cancer and knew it. This was his chance to make 
history and he was determined to make the most of it, even if it 
meant fining and closing companies and arresting executives who 
added so much as a dime to the price of their products. 

Never mind that the monthly inflation rate was back to double 
digits by the end of a year, or that a few suppliers found ways to 
bypass the rules, either because they imported raw materials whose 
prices were quoted in unfrozen dollars, or were more creative about 
assessing special “freight” charges or other new fees, or, better yet, 
had a monopoly on the market. But for a while the plan worked, 
Funaro was a hero, and the demand for products from a euphoric 
country surged. 

Our Ipiranga plant manufactured two distinct lines: mechanical 
products and electronic products. The former included dishwashers, 
mechanical scales, meat grinders, dough mixers, and cold-cut slicers; 
the latter consisted of electronic scales and printers that retailers used 
with them. The electronic products were a problem. We purchased 
many of the components from importers, so our costs continued to 
rise. But since the price for our scales was a matter of record, it 
would have been easy for any customer to lodge a complaint if we 
increased our prices. Our hands were tied. 

Before the economic plan, we had been selling up to 300 scales a 
month, for revenues of $3 million a year. After it, sales shot up to 
500, 700, then 1,200 scales a month, but since our price was fixed 
but that of our suppliers’ wasn’t, we were losing money with every 
sale. And that wasn’t all. The plant was unable to keep up with the 
demand; orders piled up, our inventory became unbalanced, and 
hundreds of scales would sit in the production area, waiting for a 

122 A Ricardo Semler 

single microchip that just wouldn’t arrive, no matter how we 
implored suppliers. And on top of that, 10 percent of our scales were 
being returned to the plant for quality problems. Customers would 
plug them in at stores hundreds or thousands of miles away and 
watch them sputter and die. 

What all this meant was that we were rapidly losing control of the 
business. That made the Ipiranga plant an ideal candidate for the 
amoeba treatment. We decided to segregate the electronics side of 
the business, moving it to a new plant at Jabaquara, ten miles away. 
To run the new unit, which we called DBData, we chose a person 

named Jose Joao Fiasco. (I know. But what could we do? That was 
his name.) 

Fiasco was a heavy, well-dressed man with a high IQ and impecca¬ 
ble manners. He had been born to an upper-middle-class family and 
had both an engineering degree and an MBA. He had been a 
successful sales administration manager at Santo Amaro, but had no 
management experience or any direct contact with workers. He was, 
however, a true believer in Semco’s policy of trusting its employees. 

His top lieutenant was Rogerio Ottolia. I had met Rogerio two 
years earlier, when he managed a plant Clovis and I were considering 
buying. Rogerio impressed us with his enthusiasm and unflappability 
in the face of our persistent questions about engineering, finance, 
and sales. The deal didn’t work out, but we hired Rogerio first 
chance we got. An electronic engineer by trade, Rogerio had a loping 

walk and bright, dark eyes. He would prove to be a never-ending 
source of innovation. 

Off they went, packing up the machinery and inventory and 
setting out with thirty other employees to their new, 15,000 square 
foot shop. Once they settled in, the mostly young work force, some 
of them electronic experts, created an entirely new culture. The 
Kids, as they soon were known, culled through the inventory and 
discovered, among other items, three years' worth of expensive 
integrated circuits for a scale that was no longer in production 
several hundred displays that didn't fit the housing of the scale they 
were making, and hundreds of transistors that didn't work on any 
scale but were somehow never returned to the supplier. The villain 
was MRPS, which had been ordering some items at double and triple 

Maverick a 123 

the proper rate. Not wanting to risk a repetition, the Kids set up a 
just-in-time inventory system and installed a Brazilianized kanban, 
based on the Japanese system in which components and raw materials 
are color-coded according to their use in the assembly process and 
stocked near the machines, so workers can immediately see when 
they are running low. 

Soon sales at the new unit were adjusted to meet its production 
capacity, and the defective scales were recalled and fixed. In just a 
few months, DBData became Semco’s flagship operation, with pro¬ 
ductivity twice that of the food service equipment plant, inventories 
reduced by 40 percent, and defects down to less than 1 percent. 

What caused this success? The Kids innovated all over the place. 
Each day started with a short meeting attended by all the plant’s 
employees, who wore white coveralls, Japanese-style. Financial infor¬ 
mation was regularly posted on the bulletin board, and an open office 
plan encouraged easy access to everyone by everyone. My favorite 
innovation was a board at the plant entrance with the name of each 
employee and next to it a wooden peg. As each person arrived in the 
morning he would hang one of three metal tags on the peg: a green 
tag stood lor “Good Mood,’’ a yellow tag for “Careful,” and a red tag 
for “Not Today—Please.” Maybe it was cute, but the Kids took it 
seriously, selecting their tags carefully and paying heed to those of 

Incidentally, the green tags usually predominated. 

The remaining employees at Ipiranga were divided into two units: 
those who made dishwashers and those who made the slicers and 
peelers and other food preparation products. The mood improved 
there, too, especially on the dishwasher side, where assembly line 
workers—going the slicing team one better—altered the production 
system, stocking the components on the shop floor and working in 
teams that moved about the plant as needed, rather than remaining 
in the same spot. A welder might assemble machines one day, then 
drive a forklift the next. Similarly, machinists would change from 
tool to tool, depending on the need. 

The skeptics had always argued that we now would need many 
more people and machines when the plants were divided. In fact, 

124 A Ricardo Semler 

each of our three new units had its own plant management, sales 
staff, and financial department. But when it came to the work force, 
we were surprised. Make that astonished. We had asked each of the 
three general managers to choose the people they wanted in their 
units, much like kids choosing sides for a playground baseball game. 
Iotti, who was taking over the dishwasher unit, and Wilmar Fagundes, 
our computer expert who at this point was heading the mechanical 
products unit, made out their wish lists. We added them up, then 
included the people who went to work with Fiasco at DBData, and 
realized that there were still a dozen or so workers who hadn’t been 
spoken for. Each manager planned to use more outside subcontractors 
and have employees double up on skills and tasks, so they would 

work more efficiently. (Most of the surplus workers were retrained for 
other jobs.) 

Quality also rose at the two Ipiranga units, as it had at DBData. 
And all three units eventually began delivering products the next 

day, which was unheard of in Hobart’s history, or in Semco’s for that 

Rather than creating waste by breaking up our plants, we were 
eliminating it. My guess is that as Ipiranga grew bigger and more 
complex, there were more nooks and crannies where marginal or 
unproductive employees could hide. The paper shufflers multiplied, too. 

How big is too big? The British author Antony Jay, in his book 
The Cor Poration Man , reminds us that we have been hunters for five 
million years, farmers for nearly 300 generations, and industrialists 
for a relative blink of an eye. Through virtually all of human 
existence we have been part of small groups, usually five to fifteen 
people. How can a corporation ignore so much experience and expect 
its employees to adapt to groups of 1,000, much less 10 000 > 

By all means hire 10,000 workers, if you have the products and 
the markets. Just organize them in small business units. A company 
can be gigantic with 1,000 employees if they are all under the same 
roof. Similarly, there are small companies with more than 50,000 
workers, but you won’t find more than a few hundred working 
together. The truth is, there isn’t a program at Semco that won’t 
work at a company with 5,000 or 10,000 or 100,000 employees, so 

Maverick a 125 

long as gigantism is contained by breaking up the work force into 
small units. 

How small is small enough? For some companies the magic 
number is 500. For others, the maximum might be a few dozen. 
Usually, though, people will perform at their potential only when 
they know almost everyone around them, which is generally when 
there are no more than 150 people. That is our experience, anyway. 
Then again, we only had 200 employees at Ipiranga before we split 
up the plant and interoffice mail would take two days to move from 
one department to another, a distance of less than 300 yards. 
Gigantism, again. 

Yes, some factories are harder to divide than others. You can’t take 
a piece of an auto assembly line and move it across town. The 
challenge is to find the correct criteria for each unit—by product, by 
market, by machine, whatever. It the equipment is impossible to 
move, then divide the existing factory in two, but make sure there 
are different people in each unit responsible for such areas as sales, 
marketing, production, finance, human resources, and all the rest. 

In times of robust economic growth we have found our divided 
plants make more money than they did when they were larger. And 
we have also found that smaller plants bounce back from bad times 
or a crisis much faster than larger ones. 

From all this I have come to believe that the economies of scale is 
one of the most overrated concepts in business. It exists, of course, 
but it is overtaken by the diseconomies of scale much sooner than 
most people realize. 



The Santo Amaro plant was the next candidate for division. We 
decided that, with its healthy backlog of orders, the marine products 
unit had outgrown its allotted space and needed to find new 

We began our search in the traditional way, leaving everything to 
the plant’s general manager and keeping our plans secret, to avoid 
frightening the employees. The manager contacted several real estate 
agents and they soon found several interesting buildings. The trouble 
was, they were in places that many of our workers, especially those 
who depended on public transport, would have a hard time getting 
to, so we stood to lose a large number of them. 

Then we realized that we were conducting the search in a manner 
contrary to our new principles. We weren’t being participative. So we 
assembled the unit’s 120 employees and gave them all the pertinent 
facts—the reason for moving, the budget for the new space, and 
everything else we could think of. Search committees were formed 
and by the next Monday the workers had come up with dozens of 
suggestions for a location. 

A few weeks later we closed the plant for a day and everyone piled 
into buses to visit the three most likely possibilities. Then, at a series 
of assemblies, the workers picked their next workplace—a vacant 
factory not far from the Santo Amaro plant, on a street called Nacoes 

128 A Ricardo Semler 

It was an excellent choice in every respect but one: it was next to a 
factory that was frequently on strike. No one in our management 
wanted front-row seats to this labor-management struggle, but we 

moved in anyway. We never considered overriding our workers’ 

Having found their new plant, our workers didn’t stop there. They 
designed the layout, in the process pushing us all another step 

Instead of a series of lathes and then a series of welding operations 
and so on, all in a long line Henry Ford—style, the workers formed 
small groups of different machines. The idea was to have, at each of 
these clusters, a team whose members would fashion a product from 
beginning to end, giving them accountability for the product’s 
quality and the enormous satisfaction that comes with completing a 
task. What’s more, these workers would know how to operate all the 
machines in their cluster, not just one, and do whatever else was 
needed, too, even drive forklifts to and from the storeroom. This 
type of organization, which had also been adopted by the workers on 

the dishwasher line at the Ipiranga plant, was known as a manufac¬ 
turing cell. 

Frederick Winslow Taylor wouldn’t have approved. Before there 
was a Henry Ford, there was a Frederick Winslow Taylor. In fact, 
there might not have been a Henry Ford without a Frederick 
Winslow Taylor. He more than anyone was the Godfather of the 
modern factory, in which thousands of nameless, faceless drones carry 
out unrelentingly repetitious tasks under ever vigilant supervision. It 
was Taylor’s belief that workers were most efficient if their jobs 
consisted of a small number of distinct movements, all scientifically 
choreographed to fit their anatomy. Taylor broke down complicated 
manufacturing processes into numerous such tasks, each of which was 
to be assigned to a set of workers. This segmentation and specializa¬ 
tion of labor, all carefully codified through rigid job descriptions 
was, Taylor maintained, the indisputable key to maximum produc¬ 
tivity. And so one worker would shovel coal out of a bin into a pile 
and another would shovel it out of the pile, carry it a few feet and 
dump it into another pile, and still another would shovel it from that 

Maverick a 129 

second pile onto a conveyer belt. And all three labored under the 
watchful eyes of a foreman. Multiply this grim little tableau by 
hundreds or even thousands of workers and you have today’s gigantic 
temples of mass production. 

Nor are the stark lessons of Taylorism confined to factories. A few 
years back, the late, unlamented Eastern Air Lines was sold to Texas 
Air, a company many times smaller. What caused this strange 
inversion:' Taylor and his job descriptions, in my opinion. For years, 
Eastern dominated many of the best routes on the East Coast. It grew 
steadily, but as it did Eastern’s pilots, flight attendants, and 
ground service personnel began to insist on detailed job descrip¬ 
tions. With the support of their unions, the airline’s employees 
strictly limited themselves only to what was in these descriptions, 
refusing to do anything else. The idle time of a baggage handler 
did nothing to alleviate the temporary lack of a ticket agent or 
maintenance worker during a crunch. Eastern found itself having 
to hire more employees, while workers already on the payroll 
weren’t always busy. 

By contrast, Texas Air hired nonunion workers who were willing 
to perform all sorts of tasks. On days off, a pilot might sell tickets; 
during peak hours a flight attendant might handle baggage. Texas 
Air took ofl and eventually swallowed Eastern, a much bigger 
company that had become inefficient and overloaded with employees. 
(I don’t mean to suggest that Texas Air didn’t have its own problems, 
which I believe were chiefly the result of its failure to make workers 
feel more involved in the company.) 

Whenever I find myself in a conventional factory, I am reminded 
of how little we have advanced since Taylor. He still intimidates 
students of management. (Maybe it’s his imposing name. Perhaps 
people would feel freer to question his teachings if they remembered 
that he didn’t study at MIT or the Stanford Business School.) I 
believe Taylor’s precise job descriptions limit workers’ potential and 
constrain the possibility of job enrichment, which dampens their 
motivation. Just think how much better job descriptions would 
be if they included not only what employees do but what they want 
to do. 

130 A Ricardo Semler 

“My dear sir,” Taylor would undoubtedly ask if he could, “isn’t 
your factory going to become a humongous mess?” 

In the terms Taylor is familiar with, probably yes. He would never 
condone a system in which, depending on what needed to be done, a 
person could be a lathe operator or a grinder, assemble the final 
product, maintain the machinery, drive a forklift loaded with sup¬ 
plies, help clean the work area or even paint the walls of his corner of 
the plant. Taylor would be appalled if workers started to make 
thingamajigs and screw together whatnots entirely on their own. Or 
if they discovered that giving suggestions, innovating, and whistling 
at work are not forbidden. 

In such a system the driving force of productivity is motivation 
and genuine interest, not predetermined routines and hulking foremen. 
That’s the difference between Taylorism and Semco’s manufacturing 
cells. We don’t believe in fragmentation. We want workers to under¬ 
stand that they are part of a whole. And we want them to figure out 
the best way to do their jobs. They’re probably going to find 

more efficient ways of doing them than Taylor or his followers ever 
would have. 

From our food service equipment and marine products units the 
manufacturing cells spread throughout Semco. I assume these layouts 
were designed by the factory committee in consultation with our 
engineering staff. But I don't really know for sure. It's been a 
spontaneous process, different at each plant. 

But if these cells made workers happy, they were giving some of 
our fiscal types heartburn. The cells sawed off the branch on which 
these sparrows of salary systems were perched, condemning these 
analysts to one hundred years of solitude. Top managers almost 
always prefer the tranquility of traditional methods to the risk of 
apparent disorganization that will rain down on any company that 

slowly frees its employees from working conditions based on narrowly 
defined parameters. 

In their squeamishness our managers were no different from some 
union leaders, who wasted little time reminding us that the workers 
all had job descriptions that precisely determined their pay. (I guess 
that made them Taylorists, too.) What would Semco's multidimen- 

Maverick a 131 

sional workers receive? Would they be paid like welders or forklift 
drivers or machinists? Naturally, the union wanted us to pay them as 
if they were in the most lucrative category all the time. Even if 
someone only spent 30 percent of his day machining, and the rest of 
the day driving a forklift, he would be considered a higher-paid 

We disagreed, of course. But keeping a minute-by-minute log of 
what a worker was doing was out of the question. Instead, we 
developed a “basket of jobs” approach: every year workers reflect on 
their various duties and the approximate time they spend doing each 
of them. Then it’s just a matter of consulting a salary survey and 
doing the math. As jobs and activities change, their individual 
formulas change. 

Some managers had another worry about the cells. They thought 
that if we let our workers take over their factories, they would never 
let us install any machinery that might eliminate jobs. But our 
workers knew what labor-saving machines could do for our ability to 
compete, and we’ve had many occasions where a factory committee 
will actually promote new equipment they think we need, even at 
the cost of some jobs. 

Workers in our cells have a way of managing more and more of the 
manufacturing process. Take quality control. We used to have 
separate departments at each unit to evaluate our products. But with 
time our workers took over this role, enabling us to eliminate jobs. 
Workers also recruit and expel new members of their teams. Today, 
anyone who applies to be a machinist at Semco will be interviewed 
by a group of machinists, not an executive, which is the worst thing 
that can happen to him, because he might be able to talk his way 
past a manager but not people who know everything there is to know 
about being a machinist and who may one day be his co-workers. If 
they say yes, the candidate has the job. I’ve not heard of a case in 
which shop-floor workers have opposed someone and a manager has 
gone ahead and hired him. I can’t imagine a Semco manager taking 
such a risk. Nor do I know of a case in which someone is approved 
by everybody on the shop floor but rejected by the manager. In 
our system it isn’t possible for one person to overrule everyone 

132 A Ricardo Semler 

Can I definitely say our manufacturing cells have made Semco 
more profitable? Some of the pluses and minuses have been tricky to 
tally. True, our manufacturing cells don’t respect the so-called 
economies of scale. Instead of buying, say, twenty machine casings at 
a time, we tend to buy casings in much smaller quantities, since 
many items are stocked right in the cell and there’s no space for 
extras. That is more expensive than buying in bulk. 

Then again, maintaining large inventories of casings can tie up 
capital. Our inventory levels have fallen to ridiculously low levels 
with the cells, and each year we do away with more and more 
stocking space. Indeed, some of our units turn over their complete 
inventory seventeen times a year, as against an industry average of 
slightly more than three such rotations. 

In some cases, it takes cells longer to make a product than it 

would on a traditional assembly line. But delivery times have still 

fallen. And there are all those quality control departments we don’t 
need anymore. 

More importantly, it is clear workers in our cells have much more 
interesting jobs than those who mechanically repeat the same simple 
set of tasks the whole day. And the cells have made people work 

much more closely, so our plants are much more finely tuned. This 
has translated into more productivity. 

Meanwhile, back at Santo Amato, things were looking up, too. 
Henrique Pinto and the employees at his mixer unit now had the 

whole plant to themselves, and were becoming more organized and 
more productive as time went by. 

There was a backlog of orders for biscuit machinery they made 

that translated into two years of production. We had already spent 

four years and well over half a million dollars to get into that 

market, and were now seeing the return on our investment, much to 

the disappointment of those who considered our entrance into the 
business a mistake. 

The unit needed more office space, better ventilation, and a new 
kitchen. Given the rosy cash flow projections, it seemed prudent to 
invest more in the plant. Anyway, the expense would have been far 
greater had not our workers done much of the work themselves The 

Maverick a 133 

ventilation system was designed by the Semco-Flakt engineers. The 
kitchen was supplied mainly by the Hobart unit and installed by a 
company in partnership with them. And most of the construction 
and painting were performed by the workers at Santo Amaro. It was 
just another example of all the good things that can happen when 
you reject Taylorism. 




All of us were ready to retire. Or better yet, become consultants. 

We thought we deserved it. We had unleashed an extraordinary 
force and were overwhelmed by the results. With prices under 
control throughout Brazil, 1986 had been a good year for Semco. 
With our plants divided, autonomous, invigorated, and at labor 
peace, and our workers more involved in their jobs than ever, 1987 
was better. 

The food service equipment plant was going great guns, especially 
the new, worker-designed assembly line for dishwashers. The Kids at 
DBData, in their white uniforms, were at least as impressive. Santo 
Amaro had 80 percent of the market for biscuit making machinery. 
And at all our plants, the factory committees, famous for their 
obstinacy at other companies, had helped us cut expenses drastically. 

Because of all this, Semco was making money—$2.2 million in 
eighteen months. I realized it was time to give everyone a stake in 
the added wealth they were creating. It was time for profit sharing. 

No more than a half dozen Brazilian companies had profit sharing 
at the time, and all of them had merely implemented off-the-shelf 
plans that unilaterally and arbitrarily doled out the rewards according 
to top management’s dictates (or whims). The companies wanted to 
motivate their workers, but at the same time retained the power to 
treat some employees better than others. This is a formula for 
resentment and division. 

136 A Ricardo Semler 

We decided we wanted a new kind of profit-sharing plan, one that 
would not only be fully comprehensible to our workers but also 
controlled by them. Before we could share the wealth, however, we 

knew we would have to share something even more valuable: 

No one can expect the spirit of involvement and partnership to 
flourish without an abundance of information available even to the 
most humble employee. I know all the arguments against a policy of 
full disclosure. Employees will use the numbers to argue for raises in 
good times, or be frightened by the numbers in bad times. Even 
worse, trade secrets will be leaked to the competition. 

Maybe. But the advantages of openness and truthfulness far 
outweigh the disadvantages. And a company that doesn’t share 
information when times are good loses the right to request solidarity 
and concessions when they aren’t. 

The origins of corporate secrecy can be traced to the insecurity of 

executives who possessed the technical skills to scale the corporate 

pyramid but weren't mature enough to handle the height. They 

wanted to be seen as different from those who had not attained their 

perch. By keeping their salaries secret, they felt they would keep 

themselves apart from others. And, of course, if they had one secret 

it was easy to make others believe they possessed more, and so were 

even more powerful, since in the modern organization power rests 
with information. 

The problem with secrets is that people usually just assume the 
worst, whether it's about profits or salaries. How do 1 know? We 
took a poll a few years earlier in which we asked everyone at Semco 
how much money they thought the company made. This was before 
we had made many changes, and our employees apparently thought 
we were unconscionably greedy. How else to explain the widely held 
belief that our profits were between 20 percent and 30 percent of our 
revenues? (Most any company would be pleased with a profit of 7 
percent or 8 percent.) After that, I didn’t need to be convinced that 
we would all be better off if our employees knew the whole truth 
When we told our managers that we wanted to share information 
about our financial performance with our workers, many were 

Maverick a 137 

alarmed. They thought the first thing the employees would want to 
know was how much they were paid. So we held our first meeting to 
discuss the monthly operating reports with a few dozen representa¬ 
tives of the factory committees and what was the first question they 
asked? How much money does a top manager at Semco make? 

We answered truthfully: $50,000 to $100,000 a year, including 
bonuses. The shock was palpable. After all, this was a country where 
the minimum wage was $1,500 a year. Never mind that none of our 
people made less than four times that. From that day on, our 
managers have been known by the factory workers as maharajahs. 
(Incidentally, the gap between Semco’s lowest- and highest-paid 
employees has grown much narrower since then.) 

The truth may not be pretty, or easily explained, but it is always 
better out in the open. We don’t tack lists of salaries on a bulletin 
board, but the information is mostly available for the asking. Yes, 
several executives requested that their salaries be kept confidential, 
and we obliged because their privacy is important, too. But it didn’t 
matter, since enough salaries were known to give everyone an 
extremely accurate idea of pay at all levels of the company. 

It was soon clear that if our executives were ashamed of their 
salaries, it might be because they felt they weren’t really earning 
them, for if they merited their pay they could easily prove their 
worth, whether it was based on specialized knowledge, experience, 
education, or the mastery of a large department with a big budget 
and staff. Executives should be proud of what they earn, and their 
salaries ought to provide everyone with an incentive to rise. 

In time our workers came to accept Semco’s executive pay and 
didn’t try to get us to lower it, as we were warned would happen. 
But they weren’t shy about letting us know when they thought we 
had too many highly paid executives floating around. 

We soon made all sorts of financial information available besides 
salaries. Of course, not everyone could understand it. Some workers 
didn’t know the difference between profits and revenue. So with the 
union’s help we began classes to teach them to read balance sheets, 
cash flow statements, and other documents. I don’t know of another 
company with such a course. 

Clovis went to the initial session and couldn’t wait to tell me the 

138 a Ricardo Semler 

first question. "What we want to know,” one worker said, "is, how 
does the company cook the books?” 

It has been rooted in the corporate consciousness that profits 
belong to those who invest the capital. Of course, this is the rule 
even at companies at which the founder originally invested very little 
and which grew largely because of the energy and talents of the 
employees. Entrepreneurs aren’t dumb. 

But some companies, looking for new ways to motivate workers, 
began to share the profits with them. This is hardly a socialist 
conceit; few ideas are as capitalist as profit sharing, which rewards 
with part of a company’s earnings the people who help generate this 
blessed surplus. Nor is it new. What is an annual bonus, after all, 
but a form of profit sharing? 

Some early profit-sharing plans were structured to return a portion 
of the proceeds generated by employee suggestions. From there it was 
but a hop, skip, and jump to plans that shared the total profits. A 
good idea became a trend, a trend became a fad, and a fad became a 
movement as companies sought a quick fix for uninterested, unmotivated 
work forces. But many just shelled out the money, leaving employees 
in the dark about how the amounts were arrived at. This undercut 
the plans effectiveness, and recent studies have found that in a high 
percentage of cases profit sharing doesn’t work. 

In 1986, Semco sponsored a seminar that brought together leaders 
of half a dozen companies with profit-sharing plans for a debate 
before an audience of 300 executives, consultants, and labor leaders 
The panel’s conclusion? There wasn't any, which was more than a 
hint profit sharing is hard to get right. 

Truth is, profit sharing doesn't create employee involvement; it 

requires it. It works only when it crowns a broad and comprehensive 

program of participation. Semco wasn't interested in having a few 

executives decide who got what. We wanted the beneficiaries to 
make those decisions. 

But which beneficiaries? By now Semco had eight autonomous 
units in four plants making dozens of products for three sectors of the 
economy. Delivery times for our products ranged from one day for a 
digital scale to two years for a turnkey biscuit factory. That’s 

Maverick a 139 

complexity. Would this special payment to our workers be based on 
the total profit made by the whole company, or the profit of each 
plant, or the profit of each independent unit, or even the profit 
attributed to each team of workers within those units? How would 
an employee from a unit that failed to make a profit and didn’t get 
an extra check feel working in the same plant as a person who 
received a large additional payment because his unit did well? Should 
the distribution formula be weighted to consider salary, length of 
service, or performance? Don’t payments that are proportional to a 
worker’s earnings reinforce the gap between the lowest- and the 
highest-paid employees? Indeed, doesn’t the distribution of profits 
represent an immense amount of money for the company but, once 
it’s divided among a thousand workers, become only a modest sum 
for each? Is it worth it? 

There were decisions to make, and the fear, especially on my part, 
that an improperly implemented profit-sharing plan could undo 
gains we had already achieved. As was our policy, we ignored rules 
and precedents and worked from reality, holding shop-floor discus¬ 
sions and meetings with factory committees and union leaders over 
the next year and a half. 

How much of our profits would we return to those who helped 
make them? That was the starting point. We weren’t about to pick a 
number out of the air, or let our workers do that, either. So we 
negotiated. We began with Semco’s total profits, the revenues minus 
expenses. Then we agreed that 40 percent would be deducted for 
taxes, 25 percent for dividends to shareholders, and another 12 
percent for reinvestment, the minimum the company needed to 
continue to prosper. That left 23 percent. 

Then the workers went to work thrashing out the distribution by 
themselves. An accounting clerk named Claudio, who represented 
the factory committee at Santo Amaro, then one of our most 
profitable plants, summed up the major sticking point at one of the 
first meetings: "Why should we share our profit with plants that 
aren’t making money?’’ he said. "It isn’t fair.” 

There was a lot of discussion on what was fair. Finally, the 
employees decided that the Semco Profit-Sharing Program (or SemcoPar) 
would work like this: Each quarter, the profit made by each 

140 A Ricardo Semler 

autonomous unit is calculated and 23 percent of that sum is delivered 
to the employees of that unit. (Of course, given the Brazilian 
economy we don’t always have profits to share, and we won’t give 
consolation prizes.) What happens to the money after that is up to 
them. They can vote to divvy it up by head count or they can 
consider years of service with the company, salary, or other criteria. 
They can decide that, rather than distribute the money, they will use 
it for some other purpose, such as for loans so workers can buy 
houses. But whatever they decide, it only applies for that quarterly 
payment. Three months later they have to decide all over again. 

That’s the theory, anyway. In reality, every Semco unit has always 
decided to split the money up evenly. That means everyone gets the 
same amount. Not the same percentage, the same amount. Someone 

making $10,000 a year receives the same profit-sharing check as 
someone making $100,000 a year. 

This is not how profit sharing conventionally works. Many compa¬ 
nies simply take a percentage and then apply it to an employee’s 
salary, which makes profit sharing more valuable for those at the top. 
Our employees have skewed the system the other way, which is fine 
with us. Now profit sharing helps balance our salary structure and 
enables us to recognize and reward those without fancy degrees who 
nevertheless contribute ten hours of their best effort every day. 

As principal shareholder—and patron—of SemcoPar, 1 have to 
admit I initially thought 23 percent was awfully high. At other 
companies it runs between 8 percent and 12 percent. But I kept 
telling myself I stood to make at least as much money in partnership 

with a motivated work force as I would as the sole beneficiary of the 
fruits of less-inspired workers. 

What would you rather have, the tail of an elephant or an enure 



We were in yet another meeting, hacking our way through an 
agenda of small, bureaucratic items like so many weeds in an 
overgrown garden, when we came to the purchase of $50,000 worth 
of file cabinets. Several departments had been waiting months for the 
cabinets and in desperation had decided to pool their requests. I 
guess they thought we would be more impressed by their plight. 

The discussion had turned to possible suppliers and prices when 
someone at the table—I don’t remember who—said, “What the hell 
are people putting in their files, anyway?’’ 

We went around the room and everyone seemed to have his own 
explanation for the Great File Crunch. “Well, of course, they 
wouldn’t be filing anything that doesn’t need to be filed,’’ was the 
majority view. I wasn’t sure. 

We didn’t buy a single new file cabinet that day. Instead, we 
decided to stop the company for half a day and hold the First 
Biannual Semco File Inspection and Clean-Out. That was the only 
way we would know for sure what all those cabinets contained. 

Our instructions were simple: We told everyone to look inside 
every file folder and purge every nonessential piece of paper. They 
were to ask themselves a question attributed to Alfred Sloan of 
General Motors: “What is the worst thing that can happen if I throw 
this out?’’ 

On the appointed day our office workers arrived in jeans and 

142 A Ricardo Semler 

overalls, ready to dig into our musty, dusty archives. 1 kept exhorting 
everyone not to be frightened about what might happen if they 
pitched one document too many. If it was really important, surely a 
customer or a supplier or someone would have a copy. 

I knew how they felt. I felt that way myself. I was one of Semco’s 
biggest file hogs, with four large cabinets and a request for two 
more. And that was on top of the two and a half cabinets each of my 
three secretaries had. 

After our cleanup, I trimmed down to a single cabinet, and that 
was pretty much how it went throughout the company. People would 
laugh when they saw six-year-old telexes, ancient catalogues, busi¬ 
ness cards with the names of people they didn’t remember and 
companies they never heard of. They also came across plenty of 
documents they desperately needed only a few months before, but 
just couldn’t find. The cleanup went so well that when everyone had 
finished Semco auctioned off dozens of unneeded file cabinets. 

That made us think more about our compulsion to collect. We 
realized that we were filing lots of documents we had no conceivable 
reason to keep, making more work for ourselves than a natural 
business required. Just how much of our clerical work, we asked 
ourselves, was similarly unnecessary? 

As our thinking evolved, we began to focus on our so-called 
support staff. I kept thinking of a story I had heard about an assistant 
cashier who had applied for work at Semco. At her interview, she was 
asked to describe her current job. "I stamp the pink copies and hand 
them to another girl,” she said. When our interviewer pressed her, 
he discovered that the woman didn’t know anything about her 

employer, except that it paid her to stamp the pink copies and hand 
them to the other girl. 

Which is why Chaplin’s Modern Times is one of my favorite movies 
The issue .s with us still. Can people truly be inspired by purely 
repetitive clerical work performed without any sense of context? How 
much of it is really necessary/ What if we could eliminate all those 
dead-end jobs and keep only positions with the potential for making 
people feel gratified? Could we run our company without secretaries, 
receptionists, and personal assistants? 1 guess I don't have to tell you 
my answer to those questions. 

Maverick a 143 

Of course, I didn’t think for a moment our executives would take 
my word for it. So 1 staged a demonstration: I sent a ten-page article 
from The Harvard Business Review to Clovis, whose office was next to 
mine. A simple thing, you think. Not quite. First, I gave the article 
to Irene Tubertini, one of my secretaries, and asked her to have it 
copied, then to bring it back to me so I could write a short comment 
on it, then to send it on to Clovis. But because the article was long, 
it first had to go to our central mailroom to be copied by a clerk who 
handled long documents. The mail is only picked up twice a 
day—between 9:00 and 10:00 in the morning and 4:00 and 5:00 in 
the afternoon. Since my test began at 11 a.m., the article sat in 
Irene’s outbasket for most of the first day. By the time it got to the 
mailroom, the clerk had left for the day, so it wasn’t copied until 
the next morning. By then it missed the morning pickup and sat in the 
mailroom for most of that day. See what I mean? Not so simple. The 
article took twenty-two working hours to travel the ten feet from my 
office to Clovis’s. 

My idea was to phase out clerical positions, redistributing their 
necessary functions among everyone else. As a first step, we suggested 
that secretaries would be more efficiently used if they didn’t have to 
wait on their bosses, serving them coffee, paying their personal bills, 
making their telephone calls. (There was a significant side benefit to 
eliminating this last task: we would end those neurotic little tele¬ 
phone wars in which the caller seeks to prove how important he is by 
refusing to come on the line until the caller is already hanging on.) 

I remember meeting with a group of about thirty secretaries. I 
told them we wanted to eliminate their jobs in a year or two and 
asked them to start thinking about what they would like to be doing 
at Semco in five years. Then we tried to push them in the 
appropriate direction, matching them with openings and sometimes 
creating new jobs for them. 

Gradually, a few receptionists and secretaries made the switch to 
marketing, sales, and even engineering. An office boy became one of 
our best draftsmen. Irene Tubertini helped start the Semco Founda¬ 
tion, working with another former secretary in the human resources 
department to research possible projects. (We eventually settled on 

144 A Ricardo Semlkr 

helping intelligent but impoverished Brazilian children attain proper 

In the end, perhaps 40 percent of our clerical workers moved on to 
other, more challenging jobs at Semco. But some liked being 
secretaries and receptionists and found similar positions elsewhere. A 
few told us they didn’t want a more demanding job because it would 
take too much time away from their families. For them we created a 
job-sharing program in which two people split a single position and, 
we all hoped, would have the best of both worlds. 

If the reaction among secretaries and receptionists to my plan was 
mixed, there was unanimity among our executives about it. They 
didn’t like the prospect of doing their own copying, dialing their 
own phone, and going downstairs to greet their visitors at the front 
desk. Some thought we were merely trying to cut expenses and 
argued that we were being pound foolish. 

“It’s going to cost you a hell of a lot more to have a highly paid 
guy like me filing my own papers, they would snarl. 

One manager came to see me armed with his Hewlett-Packard 
calculator. My secretary tells me she spends 30 percent of her time 
filing,” he said, pecking away at the tiny keypad. “Let’s say I’m 30 
percent more efficient than her.” (Never mind that she knew how to 
file and he didnt.) It still will be at least 21 percent of my time. 
Over a year that means there’s going to be $12,000 in additional 
costs to the company. And I m just talking about filing.” 

“Let’s just try it,” I kept pleading, to all of them. But the most 

effective way to make the point was to be the canary on the miner’s 
helmet. Again. 

Soon after I had taken control of the company I accumulated my 
second and third secretary, and I managed to keep all of them busy, 
too. But doing what:- 1 Filing papers I would never again have the 
slightest need for, of course. So I went from three to two to one and, 
eventually, to no secretaries. I did my own filing, which meant I 
filed fewer papers instead of fifty to sixty a week, I found myself 
filing two or three documents a month. You can take more risks with 
your own papers than anyone else can, I realized. 

Freed from most of the constraints of paperwork, I started working 

Maverick a 145 

more at home, with an answering machine, a personal computer, 
and, later, a fax. I encouraged others to work at home, too. “I need 
to be here,” some would protest. But once they tried it and found 
out how much more efficient they were without all the distractions, 
they became Work at Home evangelists, too. 

We found that in an office without secretaries it paid to reflect 
before adding to the paperwork. We still make extra copies if there is 
a chance someone will feel left out. (By the way, when we distribute 
a memo, we always list everyone getting it in alphabetical order, to 
avoid silly guessing games about prestige.) But generally, the fewer 
copies the better. And we think three times before filing anything. 
Read it, understand it, act on it, and throw it away, that's our motto 

This is not as easy as it sounds. I remember when a new manager 
brought me a beautifully bound report that powerfully and effectively 
argued against the feasibility of building a new high-pressure com¬ 
pressor for a petrochemical plant. I was impressed with his analysis 
and, after flipping through the pages and reading his recommenda¬ 
tion, looked up to tell him so, while casually throwing the report in 
the wastepaper basket under my desk. “You’re right, it’s not worth 
doing,” I said, watching the blood drain from his face until it was 
more or less the shade of our fanciest bond stationery. He had worked 
for fifteen years for a multinational corporation and wasn’t used to 
seeing the product of several long days tossed out after a few minutes 
of consideration. 

I asked him why he thought his report should be filed. 1 had, after 
all, agreed with him. He looked at me blankly for a few moments, 
then conceded with a sigh that it had served its purpose and would 
do nothing more than take up space. From that day on, he answered 
memos with terse handwritten notes in the margins. 

Another time, dozens of ball bearings worth many thousands of 
dollars were ordered from abroad, but when they arrived we discovered 
they were much too big. And the supplier wouldn’t take them back. 
It was a costly blunder, but it was difficult to know which depart¬ 
ment was responsible—purchasing, production, or engineering. The 
three department heads began exchanging memos, and inevitably the 
sniping escalated. After a few weeks of guerrilla warfare I called the 

146 A Ricardo Semler 

three of them to my office. Conspicuously displayed on my desk was 
the file of the incident, by then dozens of pages of reports and 
memos, all held together by a huge metal clip. 1 still have that clip, 
but not the reports, for 1 ceremoniously threw them in the trash 
before any of my guests could speak. 

That didn’t end all of the paper wars at Semco, but it certainly 
discouraged them. Who wants to write a bitter, incriminating 
memo, after spending hours reading files and conducting interviews, 
if he knows it is destined for the shredder? 

Alas, I knew we would never eliminate memos entirely. But I did 
find a way to make them more readable and effective and less 

One day, after a prolonged debate at one of the meetings we 
started holding on the issue of excessive paperwork, I conducted an 
experiment with a group of top managers. I began by reciting the 
following list: 

School Bus 

Broken Glass 

Then I asked those at the meeting to divide into small groups and 
write a news story based on these facts, including a headline. There 
were no restrictions on style or length. 

Thirty minutes later, everyone compared their articles. They were, 
of course, quite different. You could tell from the headlines: 

“School Children Hurt on Visit to Mayor” 

“Traffic Jam Holds Up Mayor’s Speech” 

“Mayor Throws Child Out of School Bus” 

“Mayor Christens School Bus with Champagne Bottle” 

The point is, facts can be almost irrelevant. What matters is how 
they are presented. 

Maverick a 147 

Suppose the marketing department is asked to assess whether a 
new product will make a profit. At a traditional company, it would 
issue a hundred-page report that would include market surveys, 
demographic assumptions, economic scenarios, competitor intelli¬ 
gence, manufacturing details, and more, all of which, in this age of 
What-You-See-Is-What-You-Get software, would be illustrated with 
graphs and charts. 

Trouble is, it could still be read in as many ways as there are 
readers, as the story of the mayor and the school bus proves. 

If you really want someone to evaluate a project’s chances, give 
them but a single page to do it—and make them write a headline 
that gets to the point, as in a newspaper. There’s no mistaking the 
conclusion of a memo that begins: “New Toaster Will Sell 20,000 
Units for $2 Million Profit.’’ 

And so Semco’s Headline Memo was born. The crucial information 
is at the top of the page. If you want to know more, read a paragraph 
or two. But there are no second pages. All our memos, minutes, 
letters, reports, even market surveys, are restricted to a single page. 

This has not only reduced unnecessary paperwork, but has also 
helped us avoid meetings that were often needed to clarify ambigu¬ 
ous memos. Concision is worth the investment. The longer the 
message, the greater the chance of misinterpretation. 

Of course, one-page memos took some getting used to. People 
sometimes had to rewrite them five or ten times before managing to 
synthesize their thoughts. 

This wouldn’t have surprised Mark Twain, who once apologized 
for writing a long letter because he didn’t have time to write a short 




On the day Laura de Barros joined our human resources depart¬ 
ment there was an assembly of workers, managers, and union leaders 
in our courtyard at Santo Amaro. In nine years of dealing with 
benefits, training, recruitment, and personnel evaluations elsewhere, 
Laura had come to believe union officials were seven-headed monsters— 
and workers who had anything to do with them were to be fired 
immediately. But then, she had only worked for conventional compa¬ 
nies. Now she was working for a boss who invited union leaders to 
his plant and then sat down and talked with them. 1 left a good job 
to make less money at Semco, she must have been thinking. Here I 
am working for an anarchist, not an entrepreneur. 

But it was soon apparent that Laura was a bit of a rebel herself. 
She soon became our human resources manager, working with Clovis 
to foster change and nurture daring initiatives, such as the program 
we called the Semco Woman. 

I was the one who gave her the idea. One day in 1986, about a 
year after Laura joined Semco, I put the kind of note on her desk for 
which I am famous. “What about a project for women?” it read. 
“Let’s talk when I return from vacation.” 

Corporations aren’t completely to blame for prejudice. It’s been 
inculcated in the family, at school—and probably everywhere else— 
long before anyone reaches working age. Even so, companies have a 
responsibility to combat and correct bias. Businesses, like schools 

150 A Ricardo Semler 

and parents, train people. They help individuals advance. That gives 
them an opportunity to rectify past wrongs. 

It usually doesn’t take much to evaluate a company’s commitment 
to fairness. Don’t bother with records and statistics. A look around 
the offices is often all that’s needed. If nearly everyone is white, or 
attractive, you can bet a company is biased. 

Corporations must clearly and consistently demonstrate fairness, 
and I believe exceptions can be costly. If there is even a hint that 
hiring or promotion does not depend on merit, credibility with 
workers will be threatened. Even if only one in ten workers is hired 
or promoted unjustly, who do you think people will talk about? 
Fairness is for employees like quality is for customers—it takes years 
to build up but collapses over a single incident. 

Before I got back from my trip, Laura and a colleague, Flor 
Bassanello, had developed the embryo of a program, forming groups 
of women at each of our business units. Getting them to complain 
was scarcely a problem. They wanted more money, of course—as 
muc h as men with similar jobs made. They also wanted recognition, 
at least, that they frequently worked a double shift, taking care of 
families when they got home. (I don’t have to tell you about Latin 
American machismo, do I?) 

The groups started meeting during lunch. Discrimination, lack of 
opportunity, relationships with colleagues, problems at home—there 
was so much grist for the mill. Soon there were about seventy-five 
participants, from illiterate cleaning women to managers with fancy 
degrees. The gender agenda more than compensated for their diversity. 

There is a Brazilian expression, “to have space,’’ which means to 
have earned respect, a place in the organization, the attention of 
others. Laura hoped that by creating a space for the women of Semco, 
they would begin to aspire to more than the traditional possibilities. 
And sure enough, it wasn t long before the women issued their first 
demand: they wanted Semco to modernize the locker rooms and 
office bathrooms, including the installation of bidets. This item, 
which was included in the minutes of their meetings that, in typical 
Semco fashion, were posted on bulletin boards all around the 
company, provoked howls of derision from some men. I wish I could 
say it was purely benign. But the women kept meeting, and their 

Maverick a 151 

numbers kept growing. They were serious, and before long almost 
everyone knew it. 

The first Semco womens convention occurred in December 1986. 
One hundred strong, the women boarded a bus and spent a workday 
at a hotel near Sao Paulo. 

Clovis and I opened the session. We told the women that to 
remedy discrimination it is sometimes necessary to engage in it. 
Thus, they could count on Semco for support, at least for a while. 
And the company would continue to grant them a reasonable amount 
of time to meet. But if they were going to gain power, we said, they 
would have to take the initiative and go on from there. Then we left. 

Laura later told me that at first the women were reserved and 
ill-at-ease. That was understandable. Few had been to a convention 
before, some had never been to a hotel, and several had gotten into 
arguments with husbands, fiances, parents, and bosses over their 
attendance. (This was in Brazil, after all.) 

Laura led the women through a series of dancelike exercises. The 
lights were dim, the music soft and rhythmic. The women swayed 
back and forth, not speaking, engrossed in introspection. I know 
this sounds awfully, well, touchie-feelie. But in time the atmosphere 
overpowered even the most reserved of the conferees and the women 
were laughing, crying, and hugging one another. 

Then they improvised a play in which they took parts based on 
traditional men’s and women’s roles at work. The idea was to show 
how submissive most women are, and how overbearing most men 
are, especially bosses, who were played by women with greasepaint 
mustaches and pillow pot bellies. 

That s all Laura would tell me about the convention. But the 
women apparently succeeded in convincing one another that there 
was more to be done, for from then on their movement caught on. 
Maybe even caught fire. It was standing room only at their lunch¬ 
time meetings, in spite of male supervisors who would sometimes 
find some urgent task for a woman subordinate just as a session 
was about to start. 

By this time the bathrooms had been redone, after lengthy 
negotiations over the startling cost of all that new plumbing. Their 

152 A Ricardo Semler 

next issue was more significant. Under Brazilian law mothers were 
entitled to extra money for day care only until their children were six 
months old. Then they were on their own. Few companies in Brazil 
provided more, but that did not discourage our women, who came up 
with a program and sold it to us. Semco now pays all day-care costs in a 
child’s first year, a little less in the second year, still less in the third, 
and so on until the sixth year, when children are in school full-time. 

All of us are pleased with the plan, and so are people at the many 
companies that have copied it. 

Soon enough the second women’s convention rolled around, a 
two-day meeting I’m sure was that much harder for some husbands 
and bosses to accept. But more than a hundred women took part, 
including one participant who was about to go into labor. 

It opened with an exercise called “the Wailing Wall,’’ in which the 
women scrawled graffiti on a ballroom wall specially coated for the 
occasion. The object was to let everything out. Heaven knows what 
they wrote; I wasn’t told. There was also dancing and singing, like 
the year before, but the mood at the convention was more business¬ 
like. By this time, Semco had begun eliminating clerical jobs that 
added little extra value, so the women spent part of the session planning 
training programs to help them make the transition. And after their 
return they began to evaluate all managerial positions that came open to 
see if any of them were qualified to fill them. They weren’t asking for an 
edge, just trying to make sure they would be considered. 

Some women did rise. One became a projects engineer, working 
on heavy machinery such as ship pumps and large mixing systems. 
Another, also an engineer, sold cooling towers. And several women 
pried their way into the maritime industry, a traditionally male 
domain, selling spare parts at dockside. 

But to my surprise attendance at the lunchtime meetings tapered 
off. And at about the same time the Brazilian economy hit an air 
pocket and we had to make a round of layoffs that hit women, many 
of whom were the last people we had hired, especially hard. I was 
beginning to think the project had failed. 

Laura disagreed. She believed the movement’s objectives had 
largely been accomplished and that there wasn t much more an 

Maverick a 153 

organized effort could accomplish. It was up to each woman to take 
the initiative and advance. 

Don’t worry, Laura would say. When a problem arises requiring 
collective action, the women will, too. 

About a year or so later a struggle developed between two rival 
unions. One had called a general strike across the country that, if 
successful, would have been a major blow not just to the companies 
involved but also to the other union. So it decided it would strike 
first, and sent organizers from plant to plant to stir up support. 

Semco was a favorite target. Because it was always in the news, a 
walkout at one of our plants would mean a great deal of publicity. 
The second union sent us one of its best (worst?) men: Horacio 
Pinipa, a radical more feared than admired by our workers. Horacio 
didn’t like Semco, either; he felt left out of the negotiations 
conducted by our factory committees. If he wanted to get even, this 
was as good a chance as he was likely to get. 

To this day many Semco employees still cannot explain why they 
let Horacio talk them into going on strike. (These were a different 
group of workers than at the Hobart plant.) Standing on a truck 
equipped with a powerful public address system, Pinipa went on and 
on, calling me names and trying to pit the factory workers against 
the office employees, who were resisting the walkout. At one point 
an engineer named Alberto grew so disgusted he stalked over to the 
truck and demanded the microphone. Pinipa almost hit him, then 
continued hurling insults—until he hurled one too many. “This 
bunch of women at Semco, they don’t do anything useful,’’ he 
sneered. “They’re just there to go to bed with the managers.’’ 

About forty women formed into a furious caucus in a first-floor 
conference room in no time at all. Then they spilled out the door and 
surrounded Pinipa. Pointing fingers at him, they screamed at him to 
take it back. 

Pinipa said he wasn’t going to take back a damn thing. 

Lia Guerra, an excitable Chilean, tried to take a swing at him. She 
was held back by some other women, but she kept yelling at Pinipa 
to come down off the truck and fight. 

That was when our telephone operator, Iracema, really lost con- 

154 A Ricardo Semler 

trol, went to her car, and took out the gun she kept in her glove 
compartment. (She had to drive home each night through the worst 
neighborhoods in town, and the gun made her feel safer.) Flor 
managed to talk her into putting it back, though she wasn’t really 
planning to use it. 

Pinipa, by this point cowering inside the truck cab, let us know 
he was ready to end the strike, and we guaranteed him safe passage 
to the third floor so he could sign the peace treaty. When he 
finished, Laura said she had a message for him. 

“Horacio, some of your members wish to meet with you now.” 

“Why?” the shaken union man said. 

“I think they want to exchange a few ideas with you.” 

“Tell them I can’t. I have a lot to do when I leave here.” 

W’lth that, this brave, radical union leader bolted down the stairs 
and past the women gathered at the door. 

Laura was right. This movement was only waiting. 



We spent years trying to hire Marcio Batoni. We first interviewed 
him in 1983 when, while a young production engineer at Vigorelli, 
the sewing machine company, he ingeniously developed a new 
milling machine using many components the company already had in 

To our disappointment, Marcio took another job as production 
director at a German company named Cyklop, which makes packag¬ 
ing machinery. 

We looked him up again early the next year. He came to see us 
just as he was about to leave on a month-long trip to Germany, 
during which time he would be trained at Cyklop’s expense. Given 
that, he decided it was unethical to resign upon his return. 

We are nothing if not persistent, so the following August we tried 
again to bring Batoni into the fold (though Clovis and I decided that 
this time it would be three strikes and we’re out). 

After many discussions, Marcio finally decided to join us. But he 
couldn’t leave Cyklop in a fix, so he gave four months’ notice. We 
winced, but on reflection decided we would have liked to be treated 
that way if he were leaving us. 

When he finally arrived, we immediately designated him plant 
manager of Santo Amaro. A little while later he took charge of 
Semco’s materials handling and purchasing as well. Then he became 
our engineering manager. By 1988, Batoni was general manager of 

156 a Ricardo Semler 

our refrigeration systems unit, where, despite a horrible economy, he 
managed to keep sales and profits rising. (WBen we bought the unit 
it was doing $1.2 million in sales a year; Batoni upped it to close to 
$4 million, with a 10 percent profit margin after taxes.) 

Far be it from us to leave well enough alone. Marcio had gained 
experience with our marine products while running the Santo 
plant, so he seemed a good bet for the Nacoes Unidas plant, where 
those same products were now being manufactured. 

This kind of movement, which may strike some as hectic and 
disorienting, has become a way of life at Semco. Indeed, helter- 
skelter career paths such as Marcio s have been institutionalized 
under a job rotation program in which 20 percent to 25 percent of 
our managers make a shift in any given year. I’ll bet fewer than a 
third of our people have not yet rotated at least once. 

Man is by nature restless. When left too long in one place he will 
inevitably grow bored, unmotivated, and unproductive. The cure, I 
believed, was to encourage managers to exchange jobs with one 
another. Someone in accounting, for instance, would arrange to swap 
jobs with someone in sales. They would start planning a year or so in 
advance, to give each time to learn the other’s duties and make the 
transition smooth. But as with other programs at Semco, we wanted 

employees to take the initiative. We didn’t want them to leave it to 
us to decide who went where. 

We felt that a minimum of two years and a maximum of five years 
in a job were ample. Anyone who wanted to stay put longer could, 
provided he could continually create new challenges for himself 
Otherwise, it was find a partner and dance. 

We have witnessed several successful corporate transplants that 
shocked traditionalists. One of our best managed sections, the dish¬ 
washer unit, which requires critical decisions in marketing and 
production all the time, is run by an accountant who was for many 
years our comptroller. Our electronic hardware and software products 
unit, which makes scales, is led by a mechanical engineer. Our 
current comptroller was, until recently, a sales manager. Or Batoni 
who has so far led three different business units at Semco. 

We don't like our people to be imprisoned by their degrees or 

Maverick a 157 

resumes. Nor do we let the lack of formal education limit anyone’s 
potential. We’ve had a financial director and a technical director who 
had nothing more than high school diplomas. Another technical 
director had virtually no schooling at all, but that didn’t stop him 
from leading a team of engineers from Brazil’s most sophisticated 
universities. If a scale assembler wants to be a systems analyst, we’ll 
try to make it happen. If a secretary wants to be a sales engineer, we 
will help finance her studies. 

There are so many benefits from job rotation, both for employees 
and employer, that it’s a wonder so few companies encourage it. It 
obliges people to learn new skills, which makes life interesting for 
them and makes them more valuable. It discourages empire build¬ 
ing, because people can’t very well sustain an empire if they pack up 
and move every few years. It gives people a much broader view of a 
company; a financial comptroller who constantly nags a salesperson 
about payment terms will regret it when he switches jobs and 
discovers how difficult it is to deal with a real, live customer. 

Rotation forces a company to prepare more than one person for a 
job, which is an excellent organizational discipline and generates 
additional opportunities for those who might otherwise be trapped in 
the middle of the pyramid. It encourages the spread of diverse 
personalities, outlooks, backgrounds, and techniques, injecting new 
blood and fresh vision throughout the company. It also depersonalizes 
the organization; customers and suppliers have relationships with the 
company, not Mr. Frankenheim. Similarly, it reduces the usual 
traumas when someone leaves a department or even the company, 
especially if he had tendencies toward irreplaceability—the conscious 
effort, through the inability to delegate and other related shortcom¬ 
ings, to insure that he is in absolute control. Nobody wants to pass 
on a disorganized department, nor look like a dictator or someone 
who can’t select subordinates well. 

Of course, rotation must be exercised with care, so specialists 
aren’t lost where they are needed. But this isn’t as big a problem as it 
once was at Semco, since rotation has forced us to develop more than 
one expert in many fields. 

What if someone with a unique talent decides he needs a change at 
the wrong time? Well, if he really wants a change, he’ll move 

158 A Ricardo Semler 

somewhere else no matter what we do, and then he’ll no longer be 
with us at all. We’d rather have him around to consult, even if he’s in 
a different department. But we’ve never had a case in which someone 
with a valuable set of skills wanted to move to another prison but did 
not take on the responsibility of finding a suitable replacement, even 
if it took a few years of training to bring him up to speed. They’ll 
say, “Look, I have a guy who’s been with me for three years and in 
another two, he’ll be ready to replace me.” 

And we all would wait. 

Truth is, life is pretty dizzying at Semco, rotation or no rotation. 
I rn sure our managers feel less secure than they would, say, in a 
comparable position at the subsidiary of a large multinational. Park 
yourself in a job at a huge corporation and play by the rules that are 
well known to everyone and you’ll sleep soundly at night. Semco 
managers are likely to be confronted with challenging situations all 
the time. There's no risk of boredom here. 

At times maybe Semco is too interesting. Our employees are 
always complaining that we make too many changes too quickly. We 
were always creating new programs—and I was always disappointed 
when my ideas couldn't be implemented that same week. But then, I 
have always been impetuous. I remember when, as a high school 
sprinter, I ran in an important statewide race. I had trained hard, felt 
confident, and drew the inside lane. When the starter’s gun sounded, 

I was off at full tilt, the undisputed leader. As I rounded the last 
curve my head raced on to the finish line, but my legs turned to 
rubber and refused to follow commands. Next thing 1 knew, I was 
sprawled on the track, panting for breath and watching my rivals 
cross the finish line. 1 realized then that taking first place is no big 
deal; the hard part is making it to the finish line in one piece. 

The pressure is also greater at Semco because we truly believe in 
the market. We don’t protect anybody from the vicissitudes of the 
business cycle or the crazy Brazilian economy. This is not for 
everybody certainly not for bureaucrats who spend careers digging 
themselves in like soldiers in the trenches. 

Because of this pressure, we are great believers in professional 
recycling, aka sabbaticals. We call it our Hepatitis Leave. When 

Maverick a 159 

people tell us they don’t have time to think, we ask them to consider 
what would happen if they suddenly contracted hepatitis and were 
forced to spend three months recuperating in bed. Then we tell them 
to go ahead and do it. 

Professionals—for now this program is limited to midlevel jobs or 
higher—can take a few weeks or even a few months every year or two 
away from their usual duties. They can spend the time reading books 
or articles, learning new skills, or redesigning their job. Or they can 
just think. In fact, that’s the point of the Hepatitis Leave. It’s not 
designed as a cure for overwork. It’s meant to create a hiatus in a 
career during which our people can stop and rethink their work lives 
and their objectives. It’s a point at which they can plan ahead, sort 
out priorities, and come back in better shape to tackle their goals. 
(Just so they remember to come back to us when their recuperation 
period is up.) 

But sometimes our flexibility backfires. We once spent more than 
a decade preparing an executive to be a production manager. Then we 
learned on our “What Do \bu Want to Be in Ten Years’’ form that he 
was more interested in marketing, so we paid for graduate schooling 
in that field at the country’s foremost business school and then 
transferred him. That’s how he became head of the marine products 





I had a premonition about the letter when I saw the return 
address. I was certain there weren’t any homes in that section of 

Sure enough, the message inside was spelled out in letters cut 
from magazines, like a film noir ransom note. It laid out details of a 
purported kickback scheme between two Semco executives and a 
supplier of steel plates and had been written, the anonymous author 
said, to repay a debt owed to my father. 

The letter arrived at my home just after we had placed orders 
worth more than $500,000 with the supplier it named. Since we had 
not done business with the company, and no other bidders had been 
consulted, I suppose we would have been within our rights to call in 
the executives and question them. But we had embarked on a new 
era of trust at Semco and it seemed consistent with that spirit to 
discount accusations that were not accompanied by solid evidence or 
even the accuser’s name. 

Instead of investigating the executives, Clovis and I decided to 
investigate the complainant. If the allegations were false, then we 
certainly didn’t want such a person working for us. If they were true, 
we would take the appropriate action, but we also wanted to know 
why such an underhanded method of exposure was necessary. 

The return address was handwritten, and Clovis hired a forensic 
graphologist to study the scrawl. As part of his investigation, he 

162 A Ricardo Semler 

went through our files. A week later we had his report, which made 
us glad we had handled the case with discretion. Our handwriting 
expert was sure the author was the same employee who had written a 
similar letter, this time signing his name, a year earlier. We had 
ignored it because the accuser had a history of making unfounded 
charges against supervisors. It was no secret that he felt he had been 
unfairly passed over for promotions. 

The two executives were exonerated and the accuser was dismissed. 
Only then did everyone come to know about the case. 

But before you get the idea that we take impropriety lightly, let 
me quickly tell you about another case. A few months after we took 
over the Hobart factory, a worker noticed that three parts and service 
department employees often worked late for no apparent reason. One 
night he stayed behind, crawled into a small closet, and waited. 
Watching through a tiny hole in the door, he sasv the trio sneaking 
out parts to their cars. The thieves were caught, fired, and arrested. 
They had been stealing from us for about a month. I hate to think 
how long it would have gone on without the help of that worker. 

These two cases, I believe, illustrate crime and punishment, 
Semco-style. We will always press charges when we are sure someone 
has committed a criminal offense. Always. But short of actual 
criminality—or in cases when there is reasonable doubt—we are 
extremely reluctant to get involved. We don't make written 
admonishments, suspend people, dock pay, or anything of the sort. 
We don t want to become boarding school deans. We'll talk about 
what bothers us, but generally won't do anything official. Of course 
if we feel someone is truly unfit for employment in our environment’ 

well tell him so. If he doesn’t change, eventually well fire him. But 
we don t slap wrists. 

Like so much of our thinking, our views on discipline are shaped 

by the conviction that our employees are adults, and should always 

be treated as such. As Semco changed, we came to stress two points 

about individual conduct: one, each employee is responsible for his 

own actions; and, two, what people do on their own time is their 

own business. We care only about an employee's work, not his 

private life, so long as it doesn't interfere with his performance 
the job. 


Maverick a 163 

Once, at one of those barbecues our plants throw when they meet 
their end-of-the-month goals, a few workers started smoking mari¬ 
juana. They would have been dismissed instantly at many companies, 
but not at Semco. We called a meeting with the factory committee 
and said that smoking marijuana on company property was simply 
not allowed. We knew the people involved, but didn’t mention their 
names. The object wasn’t to punish them, just to make sure it 
wouldn’t happen again. 

If we don’t feel imperiled by our employees’ vices, we don’t feel 
responsible for them either. If a worker is an alcoholic, it’s not 
Semco’s role to try to get him on the wagon. Alcoholics Anonymous 
does that. If he chain-smokes, it’s not our corporate obligation to 
clear his lungs. If he’s addicted to cocaine and wants medical or 
psychiatric treatment, we’re willing to help him get it, but—this is 
one of those Important Buts—he has to take the first step. 

This may seem unfeeling or even out of fashion in an era when 
many corporations believe—no, proclaim—that it is their duty to 
assist employees struggling with drugs or booze or personal prob¬ 
lems. They mean well, I’m sure. But we don’t want to turn our 
managers into father figures, even if it makes them feel warm and 
cuddly inside. We don’t want to be a big, happy family. We want to 
be a business. No one should ever fall for that “we’re-a-family” line. 
Ask someone who retires and three days later can’t get through the 
gate. Even a worker who gets sick is forgotten if he doesn’t return to 
the plant soon. A few friends from the company may come by at 
first, and the checks continue to arrive each month, but people in 
that awful situation soon realize that no company is a family. 

Businessmen want to go to heaven, just like everyone else. And 
what is a certain route to eternal peace if not good acts on earth? But 
the problem with paternalism is that the boss who caresses with one 
hand often slaps with the other. 

The owner of a company in our field is known as one of the most 
generous bosses in the land. When a worker’s son had a serious 
medical problem that could only be treated outside the country, the 
owner paid all the expenses, which equaled many years of the father’s 
salary, a sum he could never repay. And he didn’t have to, because it 

164 a Ricardo Semler 

was a gift, not a loan. But if as he walked around his plant this 
wonderful, caring, generous owner saw someone doing something he 
didn t like, he would fire him on the spot (with severance, of course). 
That’s the flip side of paternalism. Employees give these owners a 
lease on their soul for their working lives. That can be an expensive 

We won’t make that deal at Semco. But that doesn’t mean our 
people can’t help their colleagues. We’re not against compassion; we 
just want it to be spontaneous, voluntary, and heartfelt. For instance, 
when one of our floor sweepers, at a relatively advanced age of fifty, 
learned his pregnant wife was expecting twins, Lia, who worked in 
our human resources department, went through the factory and the 
offices and collected donations of money and children’s clothing. No 
one put her up to it; she did it on her own. And nearly everyone at 
the company contributed, which made us all feel prouder than we 
would have had our treasurer just sent a check. 

Occasionally, Semco will lend employees money, but only for 
unpredictable emergencies, such as when a worker’s house is flooded 
or a spouse gets sick. As long as we feel an individual can repay us, 
we will go to almost any extent to get him the cash he needs—one, 
two or even three years’ pay. We will then deduct perhaps 20 percent 
or 30 percent from his paycheck until we are even. But if a worker 
has an opportunity to buy the house he lives in, or a car his 
brother-in-law is unloading, we won’t lend him a cent. We don’t 
want to buy into his financial life or participate in his dreams. We 
might offer a course on budgeting, or help him calculate how big a 
mortgage he can afford, but that’s it. The love and caring we offer 
come from our people, not our policies. 

And you won't find a running track, swimming pool, or gym at 
Semco. Many companies build them to help their employees cope 
with stress. At Semco, we try not to cause stress in the first place 

Does this mean that companies should abolish all benefits and 
transform the resources they consume into additional pay? Of course 
not. The amount returned wouldn't enable their employees to buy 
the same benefits. (Heres one place where there are economies of 
scale.) So we compromise by having employees manage benefits such 

Maverick a 165 

as health care insurance. The company contributes the money, but 
the employees make the decisions on how it is spent. 

It isn’t easy living up to our laissez-faire ideals. I remember when 
Celso Violin, an employee in the human resources department, asked 
an office maintenance worker, Jose Fernandes, to feed some confiden¬ 
tial documents—payroll records, including those of a few executives 
who were still not disclosing their salary—into our shredder, which 
is so noisy it’s kept down the hall. Jose started putting the papers 
into the machine, but had to leave for a few minutes to pick up some 
trash. A few moments later a colleague rushed into Celso’s office. 
‘‘You better go out there and see what’s happening,” he said, trying 
to catch his breath. 

Celso shot into the hall and saw a security guard leaning over the 
shredder, reading the papers. This guy was trouble: he took advan¬ 
tage of his immunity as a factory committee member to cause us all 
sorts of problems. He was the sort of worker who gave the commit¬ 
tees a bad name. 

‘‘What are you doing?” Celso shouted, knowing the answer but 
not knowing what else to say. 

‘‘Reading,” the guard casually replied. 

‘‘You can’t do that,” said Celso. ‘‘They’re confidential.” 

The guard said a few nasty things and walked away. Later, he was 
called to Paulo Pereira’s office and asked why he had been spying. He 
kept saying he didn’t mean any harm. Paulo knew* what he was up 
to—he was trying to prove that some managers were making much 
more than their stated salaries. And Paulo made sure the guard knew 
that he knew. Then he asked him what he thought Semco should do 
about it. I’m sure this surprised him. But the guard had a surprise 
for Paulo. He said that, since he meant no harm by looking at the 
papers, he thought he shouldn’t be punished at all. 

I’ll bet most companies would have fired the guard instantly. 
When employees break important rules, corporations usually try to 
make examples of them, as a deterrent. But Semco is a participative 
company, and we don’t ask a question if we’re not prepared for the 
answer. We followed the guard’s recommendation and did nothing— 
except to tell him we didn’t think what he did was right. 

166 A Ricardo Semler 

* # # 

At Semco we rarely fire anyone for cause, but when we do, we 
done put a notice up on the bulletin board announcing, “Mr. Pilfer 
has decided to resign, regrettably, for reasons of health.” If an 
employee has breached our confidence, we say so. If he has left for a 
better job, we say that, too, wish him well, and sometimes express 
the hope that he will one day return. 

We always try to speak the truth and nothing but the truth. And 
on those rare occasions when the truth, for some special reason, 
cannot be told, we say nothing. We believe it is essential that all 
company communications, especially those intended for the workers 
or the public, be absolutely honest. We even apply this policy to 
journalists. Reporters from all the television networks and major 
newspapers and magazines in Brazil can talk to whomever they want 
at Semco, no matter what they have reported in the past, and do. 
And our people are free to speak their minds, without fear. 

We take another step to insure that communication at Semco 

T ^° ° f three t,me$ 3 year We distribute a questionnaire called 
What Does the Company Think?” It gives workers another chance 

to tell us whether they are satisfied with their salaries, have any 

reason to leave the company, would ever support a strike, have 

confidence in management, and so on. The results are published for 

all to see (of course), and enable us to monitor our credibility as well 
as their concerns. 

We even encourage civil disobedience in the company, though we 
do so subtly. If a request for some item gets bogged down in 
purchasing, for example, employees know that in a jam they can just 
buy it themselves and send us the bill. Workers have held protests in 
our cafeterias because they thought we should subsidize 100 percent 
of their meals, not just 70 percent. Some have refused to wear a 
uniform. Our reaction is to do nothing, except to explain why things 
are as they are. If coexistence ,s impossible, they’ll either eventually 
leave on their own or be slowly and subtly expelled from the system 
But we have people who agree with very little of what we think and 
are still here, unapologet.c and unfettered. So what if they don't weat 
a uniform, as long as they do their job. 

Maverick a 167 

A touch of civil disobedience is necessary to alert the organization 
that all is not right. Rather than fear our Thoreaus and Bakunins, we 
do our best to let them speak their minds even though they often 
become thorns in our side. 

One of the sharper thorns was a welder at our refrigeration systems 
factory, where we make cooling towers. In Brazil, people who work 
in conditions deemed to be dangerous are paid extra—20 percent, 30 
percent, or 40 percent more, depending on the risk. Our welder and 
some co-workers asked the factory committee to demand a hazard 
premium for the hot welding they were doing. But how hot is too 
hot? We told the workers we would ask a group of Semco engineers 
to inspect their working conditions, then adjust their pay accordingly. 

This seemed to please everyone but the rebel, who said he didn’t 
believe company engineers would be fair. We ignored his objections, 
and the engineers reported that there were indeed hazardous condi¬ 
tions that warranted a 30 percent increase. We immediately com¬ 
plied, making the payment retroactive to cover the weeks since the 
welders first spoke up. But the rebel, naturally, thought he and his 
colleagues deserved the full 40 percent premium. 

A few weeks later one of these same welders—the man with the 
most seniority—reached Brazil’s mandatory retirement age for full¬ 
time employees. He had been with the company many years and 
because of that had a high salary. But he was a good employee and 
we decided to rehire him as a consultant, which enabled us to 
circumvent legally the retirement rules. And, of course, we paid him 
at the same rate as before. 

Then we heard from our rebel. As the next most senior welder, he 
was hoping to inherit the salary of the senior man. We explained that 
this was an exceptional case, but he wasn’t satisfied. He went to the 
union, which said he would have to fight it himself. And so he did, 
in court. And while he was at it, he sued us over the hazard pay, too. 

This didn’t thrill us, but if a worker wants to sue us, that’s his 

A month or so later we all trooped into court. The judge began, as 
I guess he always does in such cases, by asking if the plaintiff had 
received the correct severance pay when he left Semco. 

“I don’t have it,” the rebel welder said. 

168 A Ricardo Semler 

The judge turned to Paulo Pereira, who was representing the 

company, waved a finger in his face, and said harshly, “Why the hell 
doesn’t he have it yet?” 

Because hes still an employee, Paulo replied evenly. 

"Still an employee?" said the judge, incredulous. “Anybody who 
sues a company is fired or leaves.” 

No, Paulo said, the welder had disagreed with us and decided to 
go to court, but he was still an employee. 

The judge, to everyone’s surprise, stopped the proceedings and 
congratulated Semco for its respect for the rights of its employees. 
Then the case resumed and dragged on for six months. In the end, 

Semco won both lawsuits. And the welder? He’s still with us, and 
just as sharp as ever. 



Anatoly Timoshenko was going into the arena, and the lions were 
hungry. Gathered in a meeting room at Santo Amaro was as 
antagonistic a group of people as he was likely to face in peacetime. 
If he was lucky, they would be his future subordinates. 

Timoshenko, a tall, soft-spoken man with a grizzly beard, had 
talked with all the top executives at Semco, and we had wished him 
luck. But we had been clear as a Baccarat goblet that we weren’t 
going to move a finger for him. It was up to him. 

We had hired Timoshenko several years earlier, after he answered a 
help-wanted ad. We were looking for a manager conversant in the 
legal complexities of purchasing agreements and experienced in 
delivering large orders and managing people. We also wanted some¬ 
one with a working knowledge of planning, assembling, quality 
control, cost control, finance, and marketing. 

Why such an extensive and diverse set of qualifications? The 
project we had in mind entailed the construction of entire biscuit 
factories—each consisting of between $3 million and $4 million 
worth of dough mixers, dough cutters, ovens, conveyors, electronic 
controls, the works. The thousands of components would be manu¬ 
factured at our Santo Amaro plant, then carted up and assembled 
anywhere in the world. Our customers, Nabisco, Sunshine Biscuits, 
and the like, would simply throw a switch and, voila, Ritz crackers 
and Oreos. 

170 A Ricardo Semler 

There was no one at Semco who we thought could handle the job. 
Our ad attracted three candidates, of whom Timoshenko seemed by 
far the best. He had experience with Citibank in finance and with 
several other companies in production. He was enthusiastic, too, 
until we told him that getting hired at Semco would be unlike 
anything he had ever endured. 

In these days of the new world order, almost everyone believes 
people have a right to vote for those who lead them, at least in the 
public sector. But democracy has yet to penetrate the workplace. 

Dictators and despots are alive and well in offices and factories all 
over the world. 

Most companies and employees accept this as immutable. But we 
didn t believe Semco had to perpetuate a system in which a person is 
hired who impresses his future boss but does not have the respect of 
his subordinates. Nor did we understand why we should keep a 
supervisor who wasn’t well regarded by those who were supposed to 
follow him. So we developed a program to insure that bosses were 
ratified by the people who work under them. 

Like many innovations at Semco, I don’t exactly know how it 
started. Initially, we just wanted to know why some people hadn’t 
become the successes we thought they would be when we promoted 
them, and naturally asked those who worked for them. That led us 
to draw up a form subordinates now use to evaluate their managers 
twice a year. It has about three dozen multiple-choice questions 
esigned to measure technical ability, competence, leadership, and 
other aspects of being a boss. Here are some examples: 

The subject reacts to criticism: 

a. Poorly, ignoring it 

b. Poorly, rejecting it 

c. Reasonably well 

d. Well, accepting it 

When the subject s department achieves a high level of productiv¬ 
ity, he usually: 

a. Takes credit for others’ success 

b. Gives credit to those who did the work 

c. Gives credit to the team as a whole 

Maverick a 171 

The subject conveys to his team feelings of: 

a. Fear and insecurity 

b. Indifference 

c. Security and tranquility 

The subject: 

a. Constantly reminds everyone he is the boss 

b. Occasionally reminds everyone he is the boss 

c. Rarely makes a point of being the boss 

The questionnaire is filled out anonymously, so no one is afraid to 
be honest. We weight the questions and answers according to their 
importance and calculate a grade, which is posted, so everyone knows 
where everyone stands. Seventy percent is passing, but most manag¬ 
ers get between 80 percent and 85 percent. Managers who score 
below 70 are not automatically dismissed, but a low grade usually 
creates intense pressure on an individual to change. What we want to 
see is improvement from one year to the next. Supervisors meet with 
their subordinates to discuss their grades, so the process of change 
starts very quickly. 

This employee review builds on one of Semco’s great strengths, 
our transparency. At our company people can always say what’s on 
their minds, even to their bosses—even when it’s about their bosses. 
It is instilled in our corporate culture that everyone should be willing 
to listen, and admit it when they are wrong. 

We developed the questionnaire to find out why some managers 
were failing. But we also had cases in which managers we admired— 
and repeatedly promoted—got such low marks that it made us 
wonder how we could be so wrong. We once had a man who ran a 
large department extremely well, or so we thought. On his first 
evaluation he got a 40. After looking into it, we concluded that his 
subordinates were right: he was a great salesman but a terrible leader, 
and his people were succeeding in spite of him. Solution: we made 

172 A Ricardo Semler 

him the head of a one-person sales staff, and he has shined. With a 
new leader, his former department did even better. 

But such cases are rare. Far more often the evaluation process helps 
people change. I speak from experience. My First evaluation was 
conducted by two vice presidents and my secretary, Irene Tubertini. 

I ve prepared a list of all the bad things I want to complain about,” 

Irene warned before telling me that I often didn’t let her know what 

my priorities were and then got upset when she didn’t correctly guess 

them. Her complaints prodded me to develop a system using labels 

of different colors to denote the importance of a task—red for 

extremely urgent, yellow for pressing, green for items that can wait a 

bit. (I stole the idea from the mood tags at the electronic scale 

Our next step was obvious, if radical. A manager imposed from on 
high starts with the count against him. Why not let people elect 
their boss? In a plant where everyone has a financial stake in success, 
the idea of asking subordinates to choose future bosses seems an 
utterly sensible way to stop accidents before they are promoted. It's a 
wonder it isn’t done more often. 

So we extended our evaluation system to cover proposed promo¬ 
tions to high-level positions, using another questionnaire that would 
be completed by middle managers and shop-floor supervisors. But 
since these subordinates hadn't yet worked for the candidate, we 
added a step: a group interview, or interviews, since candidates are 
often asked to return for four or five meetings. I'm sure this tries 
their patience, but it also gives them an opportunity to learn our 
business and our culture. And that gives both them and us a chance 

to find out whether we are truly made for each other before we step 
up to the altar. r 

It was Marcio Batoni who explained our strange hiring ritual to 
imoshenko. Batoni was somewhat in charge of the job opening I 
say somewhat because this is Semco, the company at which nothing 
is straightforward, and no single individual can decide whether there 
is a ,ob opening. An executive can start the process of creating a new 
job by trying to convince his colleagues that their business unit needs 
someone new. If there is a consensus in favor of adding a job, then 

Maverick a 173 

the person or people most interested in the idea put together a profile 
of the ideal candidate, listing all major qualities and requirements— 
experience, leadership ability, languages spoken—with weights at¬ 
tached to each. As a matter of corporate culture, factors such as 
academic background and personal appearance are ignored. Semco 
abounds with people who lack fancy degrees or Italian suits but are 
first-class employees nonetheless. 

The profile and a general job description are posted on all bulletin 
boards, giving our employees first chance to apply. I know what 
you’re thinking: don’t many companies do this? Yes, but here is why 
we are different: under a program w'e call “The Family Silverware’’ an 
employee who meets 70 percent of the requirements for a job will be 
chosen instead of an outsider. In other words, our people get a 30 
percent discount on a new job merely for being here. As believers in 
the power of cultural adaptation, we are willing to bet that someone 
who meets 70 percent of the requirements will quickly develop into a 
100 percenter on the job. 

Only if no qualified internal candidates come forward are outsiders 
considered. But not all outsiders are created equal. We make it a 
point to give ex-Semco employees—orphans, Laura de Barros calls 
them—an edge. Only a few have tried their luck at other companies 
or struck out on their own, but many of them have come back, and 
we have welcomed them. 

We also give a preference to friends and acquaintances of our 
employees, because no responsible person at Semco would risk his 
own reputation by recommending someone who can’t meet our 
standards. But family members are out; only distant relatives are 
permitted to work for the company, and then only in different plants. 

Only after all this fails to produce a candidate do we resort to 
newspaper ads or headhunters. We prefer three applicants for an 
opening, but will settle for two in a pinch. 

“We manufacture the ovens, conveyors, and dough mixers, ship 
them to the site, and install them,’’ Batoni told Timoshenko at their 
first meeting. “All customers have to do is put in the dough and the 
sweeteners and press the button. What we need is someone who can 
make sure every one of the thousands of components, every one of the 

174 A Ricardo Semler 

hundreds of individual deliveries, each of the dozens of critical 
engineering decisions, will be made correctly. And on time. 

“The manager of this unit will constantly be in the line of fire.” 
He looked squarely at Timoshenko. “Do you feel up to that?” 

Timoshenko thought it over for a moment and said he did. But he 
had one question about our strange procedure: Didn t we run the risk 
of hiring people who were likable rather than tough? Batoni assured 
him this wasnt the case. In the admittedly short time Semco had 
allowed employees to play a critical role in hiring their bosses, they 
have repeatedly recognized that their self-interest was the same as the 
company’s. A congenial boss might make their lives easier in the 
short run, but they knew that only the success of their department 
would insure that they continued to receive profit-sharing payments, 
or even ordinary paychecks. 

To reassure Timoshenko, Batoni told him of how three months 
earlier we had replaced our treasurer. A notice of the opening was 
posted, but no internal candidates applied. Clovis had a friend he 
thought would be ideal, a fifty-year-old treasurer at a company about 
the same size as ours. His name was Mario Fontes. But several people 
who worked in our finance department recommended Gil Ostolin, a 
thirty-year-old former Semco accountant who now ran a car dealer¬ 
ship with his brother. As an “orphan,” Ostolin got an automatic 

edge. On top of that, he had been well liked by many of those who 
were about to interview him. 

Several top Semco executives met with both men and strongly 
preferred Mario. But given Gil's popularity they didn’t think Mario 
had a chance. There was even some talk of abandoning our participa¬ 
tory hiring process just this once—it was that crucial a jol^but we 
knew our whole effort to democratize Semco would be compromised 
Both candidates were invited in to meet their potential subordi¬ 
nates on the same day. The jury, a dozen or so people who worked in 

1 nance, agreed that if neither scored higher than 80 they would keep 
looking. r 

Gil arrived first. Clovis watched as he embraced the men and 
kissed the women. What an entrance, Clovis thought to himself as 
he went back to his office to await the result. 

Maverick a 175 

Each interview took two hours. After the two candidates left, the 
questionnaires were filled out and scored. 

In the end, affection and familiarity lost out. Mario won, although 
just barely. 

After all this, Timoshenko’s interview must have seemed a huge 
anticlimax. Scoring an 84, he started work almost immediately—and 
probably started losing points right away. It was a challenging 
project, and the constant pressure to meet the tight deadlines turned 
Timoshenko into a first-class nudge. 

Semco’s first biscuit plant was delivered on time, on budget, and 
exceeded the specs required. And that was only the start. The unit 
rolled out more biscuit plants, each carrying a healthy profit margin 
of 12 percent to 15 percent after taxes. 

Soon we decided to spin off the operation into a separate unit. 
That would give it the freedom to chart its own course and insure 
that it stayed small enough to be efficient. Naturally, we wanted 
Timoshenko to run it. But that meant we would need to promote 
him to be a general manager, and under our process he would first 
have to be interviewed and approved by his future subordinates— 
including many people he had so effectively irritated. What an 
opportunity to get even. 

And it got worse. We told Timoshenko that if he wasn’t approved 
we would try to find another place for him, assuming we had one. If 
not, he would have to go, since we don’t believe in stockpiling 
talent. People get unhappy waiting on shelves. Anyway, companies 
and talents are dynamic and needs and interests change. It’s like 
keeping butter outside the refrigerator. It usually melts by the time 
you are ready to use it. 

Because of our policy of giving preference to orphans who want to 
return to Semco, we think it is healthier for people to oxygenate 
themselves in the job market than be put in some eternal hibernation. 

As Timoshenko arrived for his interview, the chatter from the 
several dozen people who had assembled ceased. Three different 
camps sat in different sections of the room. By the windows were the 
engineers, including Almir Vieira and Carlos Okamoto, the leaders 

176 a Ricardo Semler 

of the opposition. They felt Timoshenko not only had been too 
tough, but also had micromanaged, which was an anathema at 
Semco. The second group, the operation’s administrative staff, sat at 
the far end of the room, at the end of the long mahogany table. 
Many had had run-ins with Timoshenko and were against him. The 
last group, the factory supervisors, were sitting against the wall 
opposite the window. They had counted on Timoshenko to solve 
their bureaucratic problems and had seen him turn promises into 
reality on the shop floor. They were solidly in his camp. 

Timoshenko opened the meeting by calmly pointing out that 
antagonisms had been built into his former job. Yes, he had pushed 
too hard. Yes, he had made mistakes. But he would change. It was 
important to remember, he said more than once, that they all had 

There were a few nods, but most of the people just sat there in 
stony silence. After a while, Almir asked to speak. His talents as an 
electrical engineer were well known. So were his candor and temper. 
True to his reputation, he angrily accused Timoshenko of mishandling 
his assignment and placing his own ambitions first. 

A few factory supervisors tried to defend Timoshenko, but the 
engineers and the administrative people kept up their attack. As the 
afternoon wore on, it was clear Almir and some other engineers 
might quit if Timoshenko was promoted. A few resignations and the 
technological backbone of the unit would be lost. Orders for biscuit 
machinery were already stacked up. 

This time, Timoshenko repeated, he would be more flexible. 
People wouldn t have fixed positions. He wouldn’t determine who 
would run each subdepartment. The unit would run itself for a 
while, and with time the natural leaders would emerge, Semco-like. 
He kept saying that he wouldn’t be a dictator. 

Then he left the room. If the engineers prevailed, we would lose 
Timoshenko. If Timoshenko won, we might lose some engineers. 
Despite the value of these technocrats, we decided we would abide by 
the outcome. Democracy would prevail and, we trusted, Semco 
would, too. 

Well, they couldnt reach a decision that day so a second meeting 
was scheduled. Meanwhile, Timoshenko did a little last-minute 

Maverick a 177 

lobbying, going from work area to work area trying to earn a few 
more points. Some of the administrative staff seemed pliable. Most of 
the engineers wouldn’t even talk to him. 

Judgment Day arrived and the three factions took their accus¬ 
tomed places in the room. After an hour of discussion, each employee 
took a questionnaire and began to fill it in. Timoshenko gazed out 
the window, watching the others decide his fate. 

Soon enough, the results were read aloud, by group. First came 
the engineers. Interestingly, their low grades on the questions 
involving interpersonal relations were partially offset by better marks 
on questions concerning Timoshenko’s technical competence. Next 
came the factory workers. Their grades were exceptionally high across 
the board, prompting muttering from the engineers. The marks from 
the administrative people were in between. 

When the results were calculated Timoshenko had a 74—four 
more points than he needed. After more debate about the factory 
workers’ grades the group decided to eliminate the highest and the 
lowest marks on each question and recalculate the average. This time 
his score came to a 70.6. There was more griping, but everyone 
decided to respect the outcome. 

In the end, no one resigned. And Timoshenko eventually made a 
surprising selection for the job of chief engineer, the person who 
would be his right arm—the utterly flabbergasted Almir. Soon he 
and Timoshenko were working on a budget for a unit that would 
become one of our great successes. 



The changes we had made at Semco, especially the flexibility of 
job rotation and unpredictability of employees evaluating their 
bosses, resulted in people taking some strange twists and turns. Life 
at Semco was becoming more challenging all the time, as Simpliciano 
Domingos de la Sierra and Alipio Camargo, among many others, 
could attest. 

Simpliciano Domingos de la Sierra was a young man when he 
came to work at Semco, but he had already endured a lifetime of 
knocks. He left his home in a Sao Paulo slum at the age of fifteen 
and worked for some of the heavyweights of Brazilian industry, large 
multinational corporations where conditions were often deplorable 
and management always authoritarian. He soon became convinced 
that what was good for the bosses was bad for the workers, period. 

When we hired Simpliciano as a sheet metal worker Semco was 
still in the firm grasp of my father’s Old Guard. His new co-workers 
took one look at his fair skin, and saw the ease with which he 
blushed, and christened him Camarao—shrimp, in Portuguese. I 
can’t think of a nickname that was less apt. 

“I didn’t give Semco an easy time,” Simpliciano recalled. “I 
arrived at the end of 1974, and by 1975 I had already organized a 
strike for higher salaries and a cafeteria—we brought our food from 
home then, and we didn’t have much to eat.” 

180 A Ricardo Semler 

Camarao won what was merely the first round. He was the kind of 
person who slept with an eye open. Even after I took over and Semco 
started to change into a company that was more trusting and tolerant 
of its employees, Camarao was deeply suspicious. He questioned 

everything and fought with our human resources people, particularly 
Paulo Pereira. 

Sometimes, though, Camarao would stop talking just for a mo¬ 
ment and give Paulo a chance. And once in a while, they would even 
manage to exchange ideas, for underneath the hostility Camarao had 
a lively mind. Gradually, very gradually, he began to soften. Like 
Joao Soares, Oseas da Silva, and other Semco workers, Camarao 
moved from mistrust to the next stage: he listened to and thought 
about and even tried a few of our ideas. It’s understandable, isn’t it, 
that workers who come of age in an autocratic, authoritarian,' 
paternalistic environment become reflections of it. It took time for 

Camarao to adjust to the innovating, democratic, participative atmo¬ 
sphere at Semco. 

I knew Camarao had turned a corner when Paulo told me how, 
during a meeting, a union leader compared Semco unfavorably to 
Termomecamca, one of the giants of Brazil’s copper industry. “Everything 
is wonderful there," the union guy said. "There is a first-class 
cafeteria where everyone eats whatever they want, and can go back 

tor seconds. Medical care is first-class. And the salaries are among the 
highest in the region.” 

Before Paulo could intervene, Camarao jumped to our defense. 

Look, my friend,” he told the union man, “I worked for Termo- 

mecanica and I wouldn't go back there. Everything you said 

is true, but a worker there is only a worker. I'd rather earn less 

money here and be respected. Only at Semco are workers treated 

like responsible people. Everywhere else they're nothing—frisked 

like robbers. I even worked for a company where we weren't 

allowed to speak. I got plans for a part from the engineering 

department that were wrong, but I kept quiet and made the part 

anyway. If I opened my mouth, it would have been worse. I would 

have lost my job. People believe workers weren't made to think 
only to work.” 

Soon, Camarao had joined a factory committee and, although we 

Maverick a 181 

didn’t encourage it, began to oppose any arbitrary actions of his 
union. Sometimes he told us we were going too far and being too 
permissive. “Semco starts from the principle that employees deserve 
to be trusted,” he said. “But not all workers do, you know. We had a 
coffee break that began at 9:00 and ended ten minutes later. Then 
Semco bought some thermos bottles and said we could have coffee 
anytime. But some workers went for coffee and then stayed outside 
talking, when they should have been working. They abused Semco’s 
trust. The same thing happened when a telephone was installed in 
the plant. I told them it should be a coin telephone, so everybody 
would be careful using it. But Semco said the people in the office 
could use a phone for free, so the workers in the factories should be 
able to as well. But at lunch people would talk for a long time, 
especially the younger, single employees, who talked with their 

Having demonstrated an ability to see both sides of factory life, 
Camarao became a supervisor, leading more than two dozen employ¬ 
ees. The promotion was a source of immense pride, since his 
co-workers had pushed him for the job. 

But here is where the story takes an unexpected turn. “The 
workers couldn’t see that if I was measuring how long it took them 
to do something, it was because we had a delivery schedule to meet,” 
he said. “That if I asked for their attention, it was to make sure the 
product was being made correctly. I was just trying to be professional.” 

About this time there was another of those economic down drafts 
and we had to reduce our work force. As was our habit, we looked 
carefully at the middle of our organization, searching for supervisory 
levels we could eliminate. No one is insulated from such scrutiny, 
and although we were delighted with Camarao’s ascension, we felt 
his position was not absolutely essential, given the slump. So he 
went back to work at the bench, or at least tried to. Alas, the 
relationship between Camarao and many of his once and current 
colleagues on the shop floor was poisoned. They resented him, and 
the tension and hostility was so great Camarao actually asked us to 
fire him. 

Fortunately, we had a better idea: he would open a small company 
of his own and become a Semco supplier, furnishing us with sheet 

182 a Ricardo Semler 

metal parts such as casings for pumps and mixers and special shafts 
for biscuit machinery. Camarao left Semco with a bookful of orders 
and two challenges: to learn to be not only a successful entrepreneur 
but also a good boss. 

We all knew it would be hard to be one without the other. 

Alipio Camargo had a nickname, too: Dr. Disagreement. He was 
administrative manager of Hobart when Semco took it over, and, we 
quickly learned, the kind of person who would count to ten when he 
got angry—and explode by two. 

Hobart was an extremely tough company. People couldn’t cough 
without permission. So it wasn’t surprising that Alipio was a 
demanding boss. 

Under our control the plant both loosened up and became more 
profitable, but Alipio had trouble adjusting to us and to success. We 
shifted him to sales manager, but his tough-guy approach didn’t 
improve. I remember having a meeting with him to discuss whether 
we should increase prices by 30 percent. Alipio thought it was 
necessary, but not all at once, as we were contemplating. Before long 
he was screaming at me to take the job of sales manager and do 
whatever I wanted with it. Then he snatched his briefcase and 

stormed out. Certain he would be hred, he went home and started 
writing his resume. 

By then talking back to bosses was a tradition at Semco. But what 
bothered us was the way Alipio treated subordinates. He once walked 
into the office and found that a desk had been removed without his 
consent. He asked what had happened and a young woman said some 
workers had taken it elsewhere. “Who authorized it?’’ Alipio demanded, 
his anger rising like a incoming tide 

“I think—” 

You re not paid to think,” he snapped. “You’re paid to follow 

It was then that we knew we had to bring Alipio into the 
program. At least the timing was right. We had just developed the 
evaluation questionnaire, and Alipio became one of the first Semco 
managers to be rated by his subordinates. His score was about 55. 

At the follow-up meeting with his people, Alipio was told he was 

Maverick a 183 

an uncaring autocrat who would have to change. Then came a session 
with Laura de Barros. His attitude may have been appropriate at 
Hobart, she said, but now he worked at Semco. 

Alipio heard the message and started to improve, which was 
fortunate because we had in mind making him general manager for 
administration and finance for the whole company. He had all the 
technical skills, but we really should have known better, for as soon 
as the promotion went through he immediately began acting like an 
autocrat again. 

The first floor of our headquarters was a mess: papers were piled on 
the floor and some of the chairs and tables were broken. When a 
bank manager arrived for an appointment he had to jump over a pile 
of debris to get to Alipio s office. So our new general manager 
gathered everyone together and gave them a week to clean it all up. 
Anything found on the floor after that, he warned, would be thrown 
out. While he was at it, he said he wanted them to be more 
businesslike in their manner. No arguing, he ordered. 

Alipio hired an office consulting firm, which proceeded to rede¬ 
sign the layout of his floor. Just as the new furniture was about to 
arrive, he told his people about the new plan. He wasn’t prepared for 
their action. 

“You didn’t ask us,” they angrily protested. “You should have.” 

With the entire department up in arms, Alipio was forced to back 
down and give everyone a few days to study the plan and recommend 
changes. In the end they approved Alipio’s design with only minor 
changes. But now everyone was happy because they had had a voice 
in the redesign. 

At about that time Alipio had his second evaluation. This time his 
subordinates gave him more than an 80. Laura met with him once 
again and told him that while we were all pleased with his progress 
we knew he could do better. She told him he should talk more with 
his people, exchange ideas—be participative. Alipio began to hold 
monthly meetings with his staff 7 . They even went out together for 
beer and dinner. That spring they all went to a ranch for a weekend 

In the spirit of togetherness, Alipio began to think about redesigning 

184 A Ricardo Semler 

his area again. This time, he wanted to eliminate the offices. “The 
floor was full of little rooms,” he said. “We didn’t need them.” 

But Alipio had learned enough by then to know that he couldn’t 
eliminate the offices suddenly and unilaterally. So after paving the 
way with a lot of talk, he started by convincing everyone to let him 
remove a single wall from each office. When they were used to that, 
he removed another wall. And then another. 

“Everybody accepted it,” he said. “I think for the first time, we 
were a true team.” 

Alipio had become a Semco manager. The final proof was when he 
reorganized his department, bringing in computers and streamlining 
all the administrative procedures. He believed he could pare the 
number of employees in his unit from forty-four all the way down to 
eleven. Many were transferred; some were laid off. In true Semco 
fashion, he fully expected to be in the latter category. 

And, indeed, soon he had made so much progress consolidating 
our many bank accounts and eliminating so many accounting cost 
centers (once more) that the department really didn’t need a manager. 

“If you have another job for me, great,” he told me. “If not, I’ll 
move on to another company. But I’d like to stay.” 

Eventually he went back to the Hobart plant to head our food 
service equipment unit. 

The pyramid, the chief organizational principle of the modern corpora¬ 
tion, turns a business into a traffic jam. A company starts out like an 
eight-lane superhighway—the bottom of the pyramid—drops to six lanes, 
then four .; then two, then becomes a country road and eventually a dirt 
path, before abruptly coming to a stop. Thousands of drivers start off on 
the highway, but as it narrows more and more are forced to slow and stop. 
There are smash-ups and cars are pushed off onto the shoulder. Some 
drivers give up and take side roads to other destinations. A few—the most 
aggressive—keep charging ahead, swerving and accelerating and bend¬ 
ing fenders all about them. Remember, objects in the mirror are closer 
than they appear. 



Despite our efforts to simplify Semco’s structure and prune away 
managerial excesses, our executive ranks were still bloated. Some of 
us in leadership positions were all too aware of the company’s 
bureaucratic girth. So were the factory committee members, who 
constantly complained that there were too many bosses, and that 
they talked too much and were too slow to act. 

Everyone knew what needed to be done. Several years earlier, in 

1986, our food service equipment plant had provided a vivid case 

study in what is now known as downsizing. Six positions were 

eliminated: the managers of marketing, finance, production, sales 

administration, data processing, and technical assistance. (Four left 


the company and two went to work for other Semco units.) Together 
they comprised the entire second tier of the plant’s hierarchy, nestled 
right under the general manager’s nose. 

The purge probably wouldn’t have occurred as an evolutionary 
step, even at this, our most daring plant. As it turned out, though, I 
was doing my stint at the plant, so the nose in question was mine. 
My predecessor, in a classic effort to turn the plant around, had 
brought in more managers. But the workers began complaining that 
the new bosses were getting in the way, dragging down performance, 
and costing us money. 

I agreed that we could run the plant with the people in the second 
tier, and so did the people in the third tier. Still, I wasn’t about to 

188 a Ricardo Semler 

tell anyone how we were going to get along without the six 
department heads. That was—and is—basic to my style. No one can 

get me to decide a damned thing. My goal is to get people to decide 
things for themselves. 

"Who’ll do the marketing?" someone asked me after the market¬ 
ing manager, Mara Mantovani, had cleaned out her office and went 
off to work for her father s small manufacturing company. 

"Let’s not worry about it,” I said. "It’ll get done somehow.” 

And it did. Mara had four marketing people under her in the third 
tier and they just divvied up her job, task by task, all by themselves. 
There was no formality to it. No memos. No meetings. No 
approvals. Among the marketing manager’s responsibilities was to 
see to it that we were represented at trade fairs. When the next event 
approached, one of her four deputies took care of it. When the 
account executive from our advertising agency came by, another of 
the four dealt with him. Still another took care of the new brochures 
In our new arrangement, the marketing department was no longer 
headed by an individual. It had become a team. 

Oh yes, the moral of the story: far from missing the second tier 

the food service equipment unit increased sales and profits. Less was 

Even so, our other plants were reluctant to follow Ipiranga's lead 
The resistance was led, naturally, by m.dlevel managers who feared 
for the, ,obs Once you put something like this in motion, they 
undoubtedly thought, where will it stop? Bureaucracies are built by 
an for people who busy themselves proving they are necessary 
especially when they suspect they aren’t. All these bosses have to 
keep themselves occupied, and so they constantly complicate everything 
The heart of the problem is the pyramid, the basic organizing 
principle of the modern corporation. It gets narrower as it rises 
rewarding the few who keep climbing but demoralizing a far greater 
number who reach a plateau or fall by the wayside. What can be 
expected from the employees on the lower levels, whose opinions are 

rTTrhT 'l Wh ° m eXplanadons are rarel >' given? They know 

that the decisions that matter, the decisions that will affect them are 

nnade on high. Is ,t reasonable to ask, year after year, for a special 

effort from these people, and then reward them with a few public 

Maverick a 189 

thank-yous and perhaps an extra month’s salary, while the lucky few 
at the top enjoy fancy offices and shiny new cars, not to mention 
bonuses that can exceed the combined salaries of a hundred or even a 
thousand ordinary workers? 

Those who make their peace with the pyramid and develop 
specialized skills—accountants or engineers of various persuasions, 
for example—can expect job security. But their fate can be to enter 
and leave at the same time every day, doing what they have done for 
years or even decades. Is it reasonable to suppose that they will 
continue to be motivated? 

Those who maintain a smidgen of ambition expect gradual in¬ 
creases in power, responsibility, title, and money. Because of the 
constraints of the pyramid, organizations are not ready to promote 
them fast enough to satisfy them, so many firms take the easy way 
out and create an extra level or two for their overachievers. What 
harm does it do? It’s just a few more lines on the organizational 

But soon there is such a pollution of titles and levels—and a 
diffusion of responsibility and authority—that much of manage¬ 
ment’s time is spent dealing with the inevitable conflicts, jealousies, 
and confusion. Six or seven levels are common even at a time when 
flattening the pyramid has become a fad. Bigger corporations have 
twelve or fourteen tiers. Given the typical executive’s respect for 
hierarchies, how is it possible for anyone five rungs from the factory 
floor to know what’s going on there? He can’t, so he distracts 
everyone around him with memos, phone calls, and meetings trying 
to find out. 

I wanted our people to have more contact with one another. I 
wanted less clutter. I wanted fewer levels. I wanted more flexibility. I 
wanted a new shape for our organization. 

In the fall of 1988, my soon-to-be-wife, Sofia, and I rented a 
house for two weeks on the Caribbean island of Mustique. Our plan 
was simply to sit on the beach and relax. I brought along a small 
library—Thomas Mann’s Death in Venice , Machiavelli’s The Prince , 
Ibsen’s A Doll's House , Parting the Waters by Taylor Branch, Alex Haley’s 
Autobiography of Malcolm X, and a collection of poems by e. e. 

190 A Ricardo Semler 

cummings. Even so, I spent most of my time thinking about Semco. 

The progress we had made. The distance we had to go. 

I had been trying to get people to react in a new way, to make 

more decisions, to take control of their work lives, to work faster and 

think faster. But we were still wedded to an archaic structure. We 

persisted in cramming our managers into that inflexible pyramid. 

Up, down, or out. Those were the only options. There was some job 

rotation going on at Semco, but not nearly enough to compensate for 

the claustrophobia of the pyramid. It was too hard for people to float 

around the company, accumulating authority informally. The only 

way to progress was to move up a rung, even if an employee had no 

desire to take on additional managerial responsibilities. The pyramid 

also limited how much anyone could advance financially by honing 

their skills in the same job, since it was virtually impossible to pay a 

person more than his boss. Hell, the pyramid even prevented people 

rom talking directly to colleagues who were too many rungs above 

Watching the gentle Caribbean waves, it suddenly seemed so 

o vious. Why not replace the pyramid with something more fluid. 
Like a circle. 

A pyramid is rigid and constraining. 

A circle is filled with possibilities. 

Why not try to round the pyramid? 

We began sketching it out in Mustique. Sofia and I would find 
sticks and draw in the sand, stepping back to ponder the implica¬ 
tions of our handiwork. Back in Sao Paulo, I continued to refine the 
.dea. After a few months, I was playing with three c.rcles-a tiny 

one in the center and two progressively larger ones—and some 

That small innermost circle would enclose a team of half a dozen 
people (including me), the equivalents of vice presidents and higher 
in conventional companies. They would coordinate Semco’s general 
policies and strategies and be called Counselors. 

The second circle would enclose the seven to ten leaders of Semco’s 
business units and be called Partners. 

The last, immense circle would hold virtually everyone else at 

Maverick a 191 

Semco—machine operators, cafeteria workers, janitors, salesmen, 
security guards, and so on. They would be called Associates. 

And the triangles? They would be scattered around that last, big 
circle, each enclosing a single person we would call a Coordinator. 
These people would comprise the first, crucial level of management— 
the marketing, sales, and production supervisors, the engineering 
and assembly area foremen, anyone who had a basic leadership role in 
our old system. 

There would be six to twelve triangles for each business unit, and 
they would float all around the large circle, indicating that the 
people in them were moveable. Although two Coordinators would be 
on the same level of management, they could have vastly different 
skills, responsibilities, and salaries. What differentiated them would 
be their competence and leadership abilities. 

In my scheme, the smallest circle would serve as a corporate 
catalyst, stimulating decisions and actions by those in the second 
circle, the people who actually would run the company. Then would 
come the Coordinators, who would be the leaders of departments or 
specific activities, guiding teams of five to twenty Associates in their 

Movement in our circles would be freer than is possible following 
the rigid career development paths of the pyramid. Coordinators 
with relatively little ambition could stay in low gear and take in the 
scenery without worrying about crazy drivers or lengthy tie-ups. 
They could move from one job to another, assuming there was an 
opening or another Coordinator willing to trade. They could also 
decide to go back to being an Associate. And since Associates could 
earn more than Coordinators, no one would feel his paycheck 
depended on his title. A specialized software engineer who was 
nevertheless an Associate, for instance, could make much more than 
a Coordinator in the engineering department, who in a classically 
organized company would have the rank of manager and a bigger 

My circles would free our people from hierarchical tyranny; they 
could act as leaders when they wanted and command whatever 
respect their efforts and competence earned them. They could cease 

192 A Ricardo Semler 

being leaders when they wanted, or when the organization decided 
they no longer merited it. 

Because Coordinators would only be Coordinators, and could not 
embellish their titles with phrases such as "of Engineering” or "of 
Accounting, we would avoid the confusing mumbo jumbo of 
appellations and ranks common in modern corporations. And because 
there would be limits on the number of Coordinators, Associates 
would have to take on more responsibility. 

On the shop floor, in fact, each Associate would make all the 
decisions he felt confident to make by himself. If he was uncertain 
about a problem, he would consult his Coordinator. Similarly, each 
Coordinator would make all the decisions he felt confident to make. 
He would bring other issues up at a weekly team meeting presided 
over by the Partner of his business unit. This session would be held 
Monday morning, after which the Coordinators would brief the 
Associates they worked with on the results. 

Decisions that affected all our business units, such as a company- 

wi e wage increase, or decisions that one business unit did not think 

It should make alone, such as a large investment in new equipment, 

would be forwarded to another meeting on Tuesday attended by a 

representative from each unit (not necessarily the Partner), plus all 
the Counselors. 

Just three circles, four job categories, and two meetings. That's it. 

We had endless discussions about the new structure, and passed 
out circular organization charts that we asked every manager to hang 
prominently hi, office Bm for m ,„ y , th , „ drd ® 

m, eh., the, were jus, another bit of sill, symbolism from rhe 
people who brought them psychedelically painted factories. 

nrh k ,T , 1 g T g t0 put 0n m y business card?” some wondered. 
? a e at the term Partner, fearing someone might think they 

owned part of the company. "I don't see what the big deal is about,” 
still others shrugged. After all, how serious could a few circles be? 

rh Ct ' j 6 IT 5 an< ^ tnan £* es anc * tr endy new names signified 
e most ra ica c anges we had yet contemplated at Semco. Where 

there once were foremen, supervisors, and department heads, we 

would have only Coordinators—and there would be far fewer of 

Maverick a 193 

them, resulting in a startling shift in the proportion of Indians to 
chiefs. Implementing the new system meant ripping apart the 
pyramid, clearing away whole levels of management, eliminating a 
host of titles, and breaking established chains of command. Moreover, 
under the new rules one Coordinator could not report to another 
Coordinator and one Associate could not report to another Associate. 
This further flattened the hierarchy. 

Even so, the implications of the new system only became apparent 
at budget time (by which point it was too late to go back). 
Managers—I mean Partners and Coordinators—would say, “Okay, for 
this new budget I understand we have to call everyone by new 
names.” But when they tried to, they realized they were in deep 

“This foreman is now called a Coordinator,” they would begin. 
“And this supervisor, who is the foreman’s boss, is a... a... I guess 
he’s a Coordinator, too.” Pause. “But what about his boss?” 

As they were struggling with three suddenly equal people, we 
would remind them: “You know, there’s only one Coordinator for this 

“Well, why can’t we call all of them Coordinators?” 

“Because the object of this reorganization is to diminish the 

“So what happens to those people who can’t be bosses anymore?” 

“That’s up to you.” 

So they realized we weren’t just changing from Title A to Title B. 
We were reducing the number of bosses as well as the variety of 

It was a big adjustment for those who held supervisory positions 
before but now didn’t qualify as Coordinators. Suddenly, they found 
themselves floating in our new circle, sans title and automatic 
authority. But their paychecks weren’t affected, and I hoped the new, 
more relaxed atmosphere in the plant would make the transition 

Coordinators also had adjustments to make. Consider a spare parts 
supervisor who made $12,000 a year and an engineering manager 
who made $85,000 a year. Under the old system, they would have to 
communicate through several levels of intermediaries. Now, both 

194 A Ricardo Semler 

would have the same title and might report directly to the same 
Partner. Every Monday morning, they could find themselves sitting 
next to each other at the weekly meeting. Their salaries were 
irrelevant as far as their participation in decision making was 
concerned. Each had the same single vote. 

The new system also insured that a suggestion, request, or 
complaint from an Associate would instantly be heard by someone at 
the decision-making level, a Coordinator, who could take it up 
directly with a Partner at the next weekly meeting. So shop-floor 

workers who are so easy to ignore in conventional systems could get a 
response in a matter of days. 

That was the theory, anyway. But as we began to implement the 

new system, we soon ran into the problem of “ghost layers." Say we 

had promoted a machine operator to a foreman's position a few years 

ear ier. This foreman reported to a supervisor who reported to a 

departmental manager. Along comes Mr. Semler and his circles. Of 

these three managers the Partner decides to keep only one- the 

supervisor, who becomes a Coordinator. But the foreman is a good 

worker, so he is given the option of going back to his old machine, at 

his foremans salary of course, which eases some of his pain. The 

problem is that his co-workers still come to him for advice and 

ea ers ip, may e because he knows a lot or maybe because old 
habits die hard. 

We couldnt tolerate these ghost layers because we wanted to 
shorten the path people had to follow to get decisions made or 
complaints heard. The foreman who became a worker again but was 

still consulted by h.s colleagues represented an extra managerial tier 
we thought we had eliminated. 

Slowly, we ferreted out our ghosts and goblins. And we didn't stop 
here. We kept asking, “What does this department do? What does 
one o. 1 e any organization, Semco had units for finance 
administration, marketing, sales, parts, production, engineering’ 

and the rest. Each of these had ire ^ r • , ® 

, • . . ese had lts own financial manager and 

administrative manager and comptroller and on and on. And below 
each of these individuals were more managers, each of whom 
reported to the person one rung up, in the classic chain of command 

Maverick a 195 

As we put the new system in place, we would cull through these 
levels. It was like a huge game of musical chairs. There were fewer 
and fewer seats all the time. Of those supervisors left standing when 
the music stopped, some went back to their old jobs in the factory, 
some left Semco voluntarily, and some, inevitably, were dismissed. 
Their tasks were redistributed, but not formally, because we no 
longer had specific job descriptions, only those four titles. We just 
said, “This person isn’t here anymore . . and let the people involved 
figure out what to do about it, as people had at the food service 
equipment unit. 

I don’t know exactly how many leaders we lost, perhaps thirty, 
maybe more. They ranged from foremen all the way to an executive 
vice president, Arno Witte. 

“This is just semantics,” Arno had said when he first heard me 
expound on the circles. “It’s not really going to change anything.” 

Two years later he agreed with me that Semco had been so 
transformed that his own job was superfluous—he was an unnecessary 
level and he was honest and courageous enough to accept it. Arno 
was a terrific executive; it had taken us years to recruit him. Now, he 
was leaving to become president of Dunlop Ltd. 

Sadly, we lost a lot of talent in the transition. But it was a price 
we were willing to pay. Conventional companies warehouse people 
until something suitable opens up, but you know how we feel about 
that. Anyway, now that our workers could understand our balance 
sheets, there wasn’t any way to stockpile someone at $100,000 a year, 
or even $40,000. 

Semco today is characterized by an absence of structure that 
astonishes outsiders. Whenever we hire a consultant who is unfamil¬ 
iar with our company, he usually begins by asking for a copy of our 
organizational chart. When we tell him there is no organizational 
chart, we can see the sweat bead up on his brows. “Well then, we’ll 
need to make one,” he will say condescendingly. The next half hour 
is spent explaining that we haven’t used an organizational chart for 
twelve years and we don’t intend to use one now. 

The organizational chart is the birth certificate of a business, 
nothing more and nothing less. It is only useful for people who are 

196 A Ricardo Semler 

unsure about the origins of a division or a role. Organizational charts 

have their place in the modern corporation—locked away in a file 

Let those who swear by these charts take this test: Pick a 
department and ask all the people in it to classify their supervisors 
according to their proven competence and actual decision-making 
powers. Transform the results of this survey into a parallel organizational 
chart and compare it with the official version. I'd be surprised if 

they re close. The truth is, power and respect cannot be imposed in 
connect-the-dot fashion. 

I have called our structure new, but it isn't much different from 
the organization used 500,000 generations ago when man was a 
hunter. The person who saw the mammoth first became the spotter 
The one who chased the fastest after the mammoth was the Runner.' 

e one who threw the spear most accurately was the Marksman. 
Whoever managed to impose himself as the leader was the Chief. 

onsistent with this philosophy, when a promotion takes place 
now at Semco we simply issue blank business cards and tell the 
newly elevated individual: "Think of a title that signals externally 
your area of operation and responsibility and have it printed ” If the 
person likes "Procurement Manager," fine. If he wants something 

of R T’ 16 T Prmt UP Cards Sayin «> “ First Pharaoh in Chargt 
°yal Supplies. Whatever he wants. But inside the company, 

thdr nLeY ^ ° Pt ‘° nS ' <AnyWay ’ alm0st a11 cho °se to print only 

Although doubters and skeptics abounded when I first drew my 
circles, even the most cynical observers were astonished to find that 
hmgs were better off once we got t.d of the pyramid and all ,4 
ungs an ro es. Critics who months earlier told us, "We just can’t 
get by without so-and-so” were now marveling at how fast and 
efficiently our plants were operating without so-and-so. 

ra ic jams became rare. Semco was moving at the speed limit. 



“I think everyone should set their own salaries.” 

Paulo Pereira, our brilliant, iconoclastic ex-personnel manager, 
now Coordinator, was tossing another grenade. 

At another company everyone would have laughed. But at Semco, 
revolutionary notions were taken seriously. You never knew when one 
would be adopted. So when Paulo blurted out his modest proposal at 
one of our Growing Bees in the Sky sessions, everyone tried to 
imagine what such a system would be like. 

‘‘Say that again, Paulo?” Batoni requested after a few moments. 

“We’ve implemented the Round Pyramid,” Paulo said. “We’ve 
eliminated rules and cut the bureaucracy. We’ve tried to make our 
company transparent, to let our people be free. Why can’t we trust 
our employees to decide how much they should be making? Is that 
really such a big step?” 

I glanced across the table at Clovis. He had a here-we-go-again 
look in his eyes. 

Top managers at Semco received both a fixed and a variable salary, 
the latter being a bonus—from 25 percent to 50 percent of addi¬ 
tional pay a year, on average, in good years—based on performance. 
But no one was satisfied with the way these bonuses were allotted. 

Many companies have bonus systems—nothing new there. Most 
are based on profits. If the budget forecasts a $1 million profit, for 
example, a manager might earn a certain percentage of extra pay 

198 A Ricardo Semler 

when that goal is reached and more if it is exceeded. Bonuses can also 

be based on sales volume or invoicing levels. But such conventional 

formulas can be unfair and counterproductive. Say a year is 

progressing nicely and a company is closing in on 110 percent of the 

projected profit. Then a major client goes belly-up and an accounting 

adjustment has to be made, bringing profits down to 85 percent of 

the forecast. Wouldn't the suddenly bonusless managers become 

unmotivated? Or say a new tax law brings a med.ocre performance up 

to a percent profit? Aren t the shareholders going to be peeved if 
bonuses are doled out? 

At Semco we tried five or six bonus systems, without success. 
Managers would be happy one year, perturbed the next. We tried to 
tador formulas for individual plants and cut the period of the payout 
torn annually to semiannually to quarterly. Nothing worked. Then 

ZmT t a WC Sh ° Uldn,t be d ° in « anythin S w solve this 
problem. Since democracy was now part of our corporate culture, 

w y not let our managers set their own goals and, when the year was 

over, decide the extent to which they met them. From there ? t would 

a relatively simple matter to award themselves the appropriate 

bonus. And no one would dare complain about the fairness of the 

system, since the managers would determine the, own reward 

Cynics thought a few of them might take advantage and award 

hemselves undeserved or even outlandish sums, buf that dlt 

appen. ike to think it is because everyone at Semco is reasonable 

bu ' rm sure « —»—» ^. “ “ t 

greedy P “ aneS afe 3 Str0ng d ‘smcentive to be conspicuously 

had d SUC T r ^ b ° nuS system « ave Pa olo his opening He 
had dreamed of implementing self-set salaries almost from the time 

h arrived at Semco. He got the idea after attending a cou s on 

lary surveys and discovering that most systems looked empirically 
sound but actually were far from it. empirically 

Paulo had talked to Clovis and me about self-set salaries right after 
that, but we felt Semco wasn't ready. Now, two years later the few 
traditionalists left at the company were still hesitant. What’about Ifi 
t e e aborate schedules comparing salaries from job to job, company 

Maverick a 199 

to company, and industry to industry? they reminded us. Should we 
just throw them out? 

Yes, we replied. Surely there was a better way to set salaries, a way 
that would give our employees a role in a process that is always 

We were aware that, if we allowed people to set their own salaries, 
the differences among colleagues would become that much more 
irritating and disruptive. Conventional salary systems, after all, 
strive for standardization. The system we were contemplating would 
be individualistic. Executives would be called on to make difficult 
calibrations of worth and value. What if someone lacked the experi¬ 
ence of a colleague but considered himself more dynamic? If an 
executive asked for too little and got it, would he be undervalued by 
those around him? 

While Paulo and the Counselors were stewing about this, I called 
in Irene Tubertini, the last of my three secretaries. “How much 
money do you need to earn to live comfortably?” I asked her, 
watching her face flush and her brown eyes cloud over with bewilder¬ 
ment. “How much money do you need so that you will leave for 
work in the morning with the feeling that you are fairly paid? So 
that you won’t be tempted to look for another job.” 

She sat there, not quite believing what she was hearing, 
wondering what I could possibly be up to. I told her to think about 
it for a day or two, then give me a number. That would be her salary 
for the next year. Yes, I was serious, I told her. We intended to ask 
others the same question, and hold them to their answers. 

A few days later Irene told me she wanted to be paid $20,000 a 
year, which was a shade higher than she had been making. That 
seemed a little low to me, so I pushed it up by 10 percent and we 
had a deal. 

Having passed the Tubertini test, we called together those in the 
highest leadership positions at Semco and their immediate staff and 
went through the same routine. We chose these people because they 
were educated, could express their opinions, and were less susceptible 
to being manipulated by us. 

We devised an evaluation form to help them through the process. 

200 A Ricardo Semler 

They listed their age, how long they had worked at Semco, their 
current role, and how they spent their time—that is, how much of 
their workday was devoted to making decisions, meeting with 
customers, working in departments other than their own, and so on. 
Since we had done away with conventional titles and most other 
hierarchical distinctions, it was difficult to differentiate among man¬ 
agers without asking them precisely what they did. And we also 
asked our people how they would represent themselves if they were 
to leave Semco and look for a similar job elsewhere. 

After each manager completed the questionnaire, he gave it to his 

boss, who also filled it in. That gave us two sets of evaluations for 
each person. 

We didn’t want to create a process in which an employee would 
say, I want to earn $1,000 a week,” and his boss would respond 
You only deserve $750.” This wasn’t supposed to be a negotiation.’ 
We wanted each manager to focus on his role in the companv and his 
va ue. So a boss might tell a subordinate, "You consider yourself a 
veteran, but 1 think you’re still a junior purchasing manager” We 
hoped they would keep talking until they came to a meeting of le 

Only then did the subject of money come up. Before they told us 
what they wanted to be paid, we asked them to consider four criteria- 
wha they thought they could make elsewhere; what others with 
tm. ar responsibilities and skills made at Semco; what friends w h 
■milar backgrounds made; and how much money they needel to 

mm I P "n ** ** ^ ™ ^ * salary surv 

^ Price WaterhT “ TT' ^ consultant 

rl Z 7 , and C °° perS & L y brand ' F <>r the second two 
they had to look inside themselves. 

Then they decided how much they would be paid, just like that. 
Well, not quite. We had originally planned not to change a salary 

hTirr Th« ar w “ ■” h,sh - h ‘ w ” w »'»2 

h im *i, rsz r ld km “ o, " paid 

raisim? it R„, «• ! M have t0 wait anot her year before 

S c , But Smce th,s was our first, nerve-fraying trip down this 
we e t more comfortable knowing there would be an opportu- 

Maverick a 201 

nity for some give and take between a boss and subordinate before a 
number was finalized. 

We needn’t have worried. Except for half a dozen people, everyone 
set salaries that were in line with our expectations. In five of the six 
exceptions, people set salaries lower than we had projected. It wasn’t 
always easy to get them to raise their figure, either. 

“I found it difficult to define how much I should get,” said Jose 
Violi Filho, who had been recruited to Semco by his friend Iotti. 

Violi, who is barely five feet tall and weighs just over a hundred 
pounds, has one of the clearest, quickest minds I have ever come 
across. He can cut through a business problem or analyze a balance 
sheet in seconds. He came to us as an assistant accountant, rose to 
accountant, then chief accountant, plant comptroller, and corporate 
comptroller, with a stint as treasurer somewhere in there. Then he 
wanted to run a business unit and eventually became the Partner for 
the food service equipment plant. And then he went back to the 
financial area as a Counselor, doubling as a Partner for corporate 
finance. (See what I mean about careers being hectic.) But back then, 
he was a country boy, too shy and uncertain to pay himself a proper 

His supervisor at the time, Alipio Camargo, rejected the $15,000 
a year figure Violi chose. “You better redo this,” he said, “because as 
it is, we cannot even start to talk.” 

Violi raised his figure to $18,000. Alipio disagreed again, saying 
the right number was $25,000. The case was taken to Clovis, who 
decided that Alipio was right. 

Of those who participated in that first round of salary setting, only 
one sales manager awarded himself an outlandish sum. He had 
aspirations of becoming a Partner and just went ahead and set his 
salary at that level, which was $70,000 per year, about 30 percent 
more than he was making, apparently hoping the promotion would 

We felt he could become a good sales director, but not right away, 
and, since we already had a sales director, probably not at Semco. 
When we told him this, he calmly told us he was under considera¬ 
tion to be managing director at another company. He got the job, 
too, so perhaps he was right about his salary. 

202 A Ricardo Semler 

There are three reasons why reasonableness prevailed. First, every¬ 
one knew what everyone else was paid. Second, the top people_ 

Clovis, Batoni, Vendramin—are all modest about their pay. (For the 
record, I am a bit less modest; my salary reached a high of $300,000 
in the heady days of 1989, but has been as low as $120,000.) As a 
matter of corporate philosophy, we try to keep our top salaries within 
ten times our entry level pay, which is in stark contrast to the rest of 
the country, where a top manager’s salary can be eighty times more 
than a worker’s. The third reason our people tended to be modest 
about salaries has to do with self-preservation. At Semco our opera¬ 
tional budgets cover just six months, not the usual twelve. Since any 
unanticipated increase in expenditures has to be offset in a short 
period, there is little margin for maneuvering. Our people know 
salaries account for most of our operating costs, and they think about 
our six-month budgets when they set them. It’s easy to solve a 

udget problem by eliminating a salary that seems too high, and no 
one wants to stick out. 

After several years of this madness, the economic impact of self-set 
salaries has been far less dramatic than we had thought. A 10 percent 
increase has turned out to be the exception. Brazilian law doesn’t 
allow for a reduction in salary, even a self-set salary, but when 
inflation is high real income goes down when salaries are relatively 
stable. By this measure self-set salaries at Semco have gone way down 
in some years. Obviously, our people are keeping their salaries in 
check because of their concern about Semco's welfare. In good times 
or ad self-set salaries have encouraged our workers to take that 
rarest of corporate perspectives, a long-term view. And they have the 
added virtue of eliminating complaints about pay, which are always 
among a company's most contentious issues. 

About 5 percent of our work force took part in that first 
experiment. Now were up to nearly 25 percent, including most of 

our Coordinators. I don't see any reason why factory workers can’t one 
day make the same calculations. 

Now that our leaders were happy with their salaries—or at least 

zsttssp “ ,no,h "'~ — d ““* — 

Maverick a 203 

“Each of you now has the correct salary, according to your own 
estimate of your worth,” he told managers at a meeting of company 
leaders in 1989. “I propose to pay you a little less, but in return will 
give you the possibility of earning more.” 

Then he explained his new wrinkle. If Semco did well, an 
employee who agreed to risk a 25 percent salary cut—the limit— 
would receive up to 50 percent more. If Semco did poorly, he would 
suffer the 25 percent cut. So a manager who received, say, $1,000 a 
week could risk having it reduced to $750. On the other hand, he 
could, at the end of each quarter, get an amount that would bring his 
weekly salary up to $1,500. 

Employees who didn’t have high fixed expenses have taken Paulo 
up on his offer and risked more; those with nonworking spouses and 
children have risked less. The results, so far, have been rewarding to 
the players and the company, since with this program part of our 
labor costs fluctuate with profits and losses. When business is good, 
people in the program make a lot more money. When it isn’t, they 
are helping us cut expenses and lowering their profile in case cost 
cutters are called out. 



Based on our new philosophies and policies Semco now boasted 
one of the highest growth rates in Brazil. Sales, which had been 
about $4 million a year for decades, had swelled to $35 million, and 
in just a few years we went from one to six factories and from 100 to 
830 employees. 

Semco was No. 1 or No. 2 in every one of its markets; we had 85 
percent of Brazil’s marine equipment sales, 70 percent of its hydrau¬ 
lic pump sales, and 65 percent of its commercial dishwasher sales, 
and our Hobart scales had gone from 3 percent to 23 percent of the 
market in three years—this despite many competitors, most of them 
divisions of large, powerful multinationals. 

What made it all that much sweeter was that we didn’t owe a cent 
to anybody we had no debt. In fact, bank vice presidents whose 
secretaries always told us they were in a meeting when we called now 
were calling our switchboard, trying to make appointments to see us. 
And never during our expansion did we receive any help from the 

But success is more than impressive numbers in an annual report. 
This is how Oseas da Silva, a member of the factory committee at 
Santo Amaro who had once organized and led strikes against the 
company, described our transformation in Brazil’s largest newspaper: 
“It was hard to get used to. You know, like when people are used to 

206 a Ricardo Semler 

living in an authoritarian regime. When they finally let you out of 
jail, you can’t believe it’s true. The workers are motivated to work.” 

Or, as another Semco employee told a magazine: “The company 
became a paradise to work in. Nobody wants to leave.” 

It’s enough to go to your head. Which I guess it did. So early in 
1988, at the age of twenty-eight, I decided it was time to compose 
my memoirs. Sofia and I headed for our ranch in the mountains, 
where there is no phone. For nine days I wrote furiously, twelve, even 
fourteen hours at a clip, working so hard I was often surprised by the 
dawn. It all came gushing out, ten years of experimentation and 
craziness, frustration and joy. blot having much faith in electronic 
gadgets after Semco’s experience with computers, I used the word 
processor with an eraser on one end. 

I drove down from the mountains with 750 handwritten pages I 
called Turning the Tables. I sent copies to Brazil’s major publishers and 
waited. Oh, did I wait. Weeks passed. Finally, one called. I thought 
he was my best bet, since he had published both Iacocca and Morita. 

We went to lunch and I came back thinking he would make me an 
offer. He never did. 

The other publishers never got back to me, so I called them. None 
were interested in my book, although two suggested that I might 
print it at my own expense to give to Semco employees. 

That s when 1 thought of Richard Civita. 1 didn’t know him 
personally, but knew his company. Nova Cultural, a large distributor 
of magazines and a medium-sized publisher. I called him and told 
him my story. He said he was leaving tor New York that night and 
would read my manuscript on the plane. Two days later I was told he 
was willing to pay all of $2,000 for the story of my life and my 
company. Well, money wasn’t the point, I quickly told myself. I 
called my lawyer and asked him to draw up a contract, urging him 
not to push too hard, since this was the only publisher in Brazil 
willing to take a chance on me. But he is a tough man, and 
demanded a contract with a clause calling for the royalty rate to 
increase with each 100,000 books I sold. The publisher and I 

laughed about that, since we would have been happy if we moved 
5,000 copies. 

What did we know? Turning the Tables would eventually sell 

Maverick a 207 

400,000 copies and spend nearly 200 weeks on the best-seller list, 
making it Brazil’s biggest selling nonfiction book ever. The nation, 
it seemed, was eager for new ideas. As I like to say, we were sitting 
on our surfboard, a tiny speck in the ocean, when a huge wave of 
organizational unhappiness broke. We just stood up and rode it into 

I was a successful if iconoclastic manager who would soon be a 
best-selling author, and a minor celebrity. But I had another ambi¬ 
tion: I wanted to be an alumnus of Harvard Business School. 

I had started thinking about studying at Harvard halfway through 
high school. It became an obsession years later, when I was taken to 
Cambridge by friends and was impressed by what I saw. Returning 
by subway to Boston, I bought an extra token. I put it in my wallet 
with the intention of using it on the day I was admitted. 

The next year I applied to the business school, against the better 
judgment of some local counselors, who thought studying in the 
United States would do nothing for my career in Brazil. They also 
feared that a rejection, which they evidently believed was more than 
a remote possibility, would be a blow to my oversized and fragile 
ego. If I had to go to school in the United States, they advised, why 
not apply to several universities and cut my risk. 

Being thick-headed, I applied only to Harvard and anxiously 
awaited the decision. Finally the letter from Cambridge arrived. Try 
opening an envelope with your fingers crossed. 

I was sure I would be accepted. I was wrong. It took me weeks to 
recover. I didn’t tell anyone, to spare myself the added misery of 
listening to such comforting remarks as “That’s not all there is in 
life, you know.’’ From time to time, though, I checked my wallet to 
make sure the token was still there. I don’t know why, but I was 
convinced I would study at Harvard one day, if only that token didn’t 
get lost. I couldn’t cheat and put it in a safe place, either. It had to 
stay in my wallet if it was to work its magic. 

Two years later, I applied to Harvard again. This time, I was 
granted an interview at the school. I scheduled an appointment to 
coincide with a business trip and off I went to Boston. I had spent a 
good deal of thought on my attire, finally settling on the dark wool 

208 A Ricardo Semler 

suit with discrete chalk stripes I had chosen for my first day at 
Semco, a starched white shirt, and a somber rep tie. After a long 
internal debate, I decided not to wear a matching handkerchief. But 
I had devised a boldly different way to drape my overcoat elegantly 
over my arm. Having made such careful preparations, you can 
imagine my horror when I checked into the hotel the night before 
my interview, opened my suitcase, and discovered that my suit was a 
wrinkled mess. It was 11 p.m. —too late for a pressing. Then I 
remembered an old trick of my father’s. He would hang his suit near 
the shower and let the steam smooth out the material. Relieved, I sat 
back on the bed to watch M*A*S*H while my suit relaxed. Then 
came Johnny Carson. Then came sleep. The hot air in the room woke 
me with a start and I realized, horrified for the second time, that it 
was 4 a.m. and the shower was still on. I rushed to the bathroom to 

find that my suit was damper than Gene Kelly’s in Singing in the 

But what luck. It was snowing outside, and by morning my 
spongy look fit right in. 

My conversation with the dean’s assistant went well—at least I was 
impressed with my conduct. As I was about to leave, he suggested 
that I attend a class, which I did. The instructor sarcastically 
introduced me as “an elegant gentleman from Brazil,” I guess 
because I was the only one in the room wearing a suit. 

I returned to Brazil to once again await Harvard s verdict. After a 
few weeks the letter arrived. It seemed distressingly slim. I remem¬ 
ber holding it up to a light to see if entrance forms were enclosed. 
Nope. Maybe they would follow. I opened the letter. Right after the 
“Dear Applicant” I saw the word “Regrettably.” 

It would be many days before I took the letter out again and read 

it all the way through. Twice spurned! And the token was still in my 

For the first time I considered abandoning my objective. Then I 
recalled how I had forgotten my wallet in various places and it had 
always been returned, token and all. And how I had bought new 
wallets several times and carefully transferred the token. This had to 
be a sign. A Sign, I mean. 

Many months later, on a day on which I became particularly 

Maverick a 209 

irritated about losing an order, I decided to send a slightly abusive 
letter to Harvard telling everyone at that great institution exactly 
what I thought of its admissions policies. In the last rejection letter, 
I was told I didn’t have enough experience, even though I had been 
working at Semco for years. So I told them in my letter that by their 
standards they wouldn’t have accepted Steven Jobs when he was 
starting Apple. 

A few months later another letter arrived from Cambridge, and I 
braced myself for a rebuke. But the letter merely asked me to fill out 
a few forms and send in my photograph in order to be enrolled for 
the upcoming semester. What miserable timing! By then so much 
was happening at Semco that it seemed more educational to stay put. 

I finally did become a Harvard alumnus, though. A few years 
later, after I had led Semco for half a dozen years, I decided to attend 
a program for corporate executives that consisted of a month of 
intensive instruction each year for three years. I still had a lot to 
learn about formal business disciplines and techniques, but I was at 
least as curious about how the people at Harvard—the professors and 
the other CEO’s—would react to the ideas and programs we had 
been implementing at Semco. Would they think we were as strange 
as I thought they were? 

I spent dozens of hours discussing Semco during my three months. 
It would come up nearly every time a class touched on the theory of 
managerial organization or labor relations or productivity. I think I 
can safely say that I amused them with my explanation of our circular 
organization and self-set salaries. 

“Maybe I’ll try that when I get back,’’ they would say, shaking 
their heads. But somehow I got the feeling that most of them 




Factory committees with real power. Offices with open doors, and 
less than the usual number of walls. Balance sheets that are posted on 
bulletin boards. Rule books tossed in the trash. Memos never longer 
than a page. Circles instead of the pyramid. Executives that make 
any sum they say. And a best-selling book chronicling the whole 
thing. No wonder Semco was beginning to attract attention. 

My colleagues and I had made the A list. Invitations flowed in. I 
was giving fifty or sixty lectures a year by myself. My style was 
confrontational: my talk to the Association of Railroad Workers was 
entitled The Dying Railroad Industry I told the Junior Secretaries’ 
Annual Conference “How to Stop Being a Secretary’’; and the 
Convention of Financial Executives was treated to “Doing Away with 
the Financial Executive.’’ When I spoke to the undergraduates at 
Brazil’s main business school, the theme was “Why Undergraduate 
Business Schools Are Unnecessary.’’ 

I was also sought out for newspaper and magazine interviews and 
television and radio appearances. My classmates at Harvard may have 
been skeptical, but to our amazement and delight corporate execu¬ 
tives and business school professors from around the world were 
lining up at our door to get a look at the place. It got so crowded 
around our plants that some factory committee leaders complained 
that all those visitors were interfering with production. To ease the 
pressure, we instituted the Semcotour program. We would save up 

212 A Ricardo Semler 

the requests and once a month group together representatives from a 
dozen companies and take them through the company together. 
When we found we still couldn’t handle the volume, we ran tours 
twice a month and increased the number of participants to thirty- 
five. Even so, there was a five-month wait. 

It was all so flattering, so exciting. I suppose it was understanda¬ 
ble that for a while we lost track of what had brought us all this 
attention and fame—our business. In time, I cut back my lectures to 
about twenty a year, and we discouraged tours and reduced our 
exposure in the media. But it wasn’t soon enough to avoid making a 
series of mistakes. 

We were so thrilled with the attitude of the workers at the food 
service equipment unit that we fell in love with everything they 
made. The plant was in the black, but we realized too late it was 
making less than it should have been. Call it the price of vanity, 
but we lost at least $3 million over four years by stubbornly 
sticking with the unit’s mechanical scales, slicers, meat grinders, 
and potato peelers. They had about run their course commercially, 

but we kept thinking we could make improvements that would 
set them apart. 

Take the mechanical scales. They were still selling at the rate of 
1,300 a month, which represented about 20 percent of our business 
from the food service equipment unit. But the product was so 
unsophisticated it didnt have much of a future. And because it was 
being knocked off by twenty competitors, its price had deteriorated 
and there wasn’t much profit in it. So the scale didn’t have much of a 
present, either. Still, we spent hundreds of thousands of dollars 
changing the design, the materials, the suppliers, and the production 
process, frittering away managerial energy trying to fix something 
that was best abandoned. We all had such an emotional stake in 
keeping the scale alive. It had been the mainstay of Hobart in Brazil 

for half a century, and was a fixture in half the grocery stores and 
supermarkets in the country. 

We finally recognized reality and offered the manufacturing ma¬ 
chinery, patterns, and molds to any employees who wanted to start 
their own scale business. But our workers were so convinced we were 

Maverick a 213 

giving away the family jewels that it took us eighteen months to shut 
down the line and transfer the equipment. They kept forming task 
forces and pleading for time to prove that a new process or a new 
assembly method would revitalize the scale market. 

We also spent years clinging to an air-conditioning cooling tower 
for office buildings we made at our refrigeration systems plant. Our 
product was still made from steel long after the competition had 
switched to Fiberglas or plastic. Steel was stronger and fire-resistant, 
we kept telling ourselves, but it would also rust—especially in 
buildings by the sea. We spent much time and money trying to 
convince customers to accept our superiority, before switching— 
much too late—to a Fiberglas tower. 

We had trouble as well with a piece of machinery called a fluid 
coupling, which is used in large conveyor belts. We made a deal with 
a British company, Fluidrive, to jointly set up a sales force in Britain 
and Brazil and started advertising, but our line never caught on. A 
German company called Voith had a vise grip on the market and 
would cut prices every time we bid against them. Stubbornly, we 
tried to battle them head-to-head, when we should have laid back, 
biding our time. Then, when customers became tired of having only 
one supplier, we could have entered the business with a chance at 

But we were impulsive and headstrong, and it cost us money and 
jobs. Perhaps we were taking our own press clips too seriously. 

At the Jabaquara plant, meanwhile, Fiasco and Rogerio Ottolia 
had worked miracles. Productivity soared. Labor relations were 
exemplary. Sales of their electronic scales had grown from an average 
of 150 to 600 or 700 a month, and they had peaks when they sold 
more than 1,000. Our market share rose from less than 10 percent to 
more than 35 percent, and our profits rose along with it. It was 
marketing heaven. 

There was no second guessing the move from Ipiranga. The Kids 
at Jabaquara had ended the plague of late deliveries and solved the 
unit’s chronic quality problems, among other things closing an 
opening in the scale s housing that enabled cockroaches to enter. 

214 A Ricardo Semler 

(Once inside, they would eat all the bits of food they could and then 
move on to the integrated circuits.) 

When we saw we were making a bundle, we did what any shrewd 
company would do and invested heavily in the new plant. Then we 
hit another of those economic air pockets for which Brazil is famous 
and everyone suddenly stopped spending money. Our distributors 
were loaded with electronic scales, and sales dropped to less than 200 
a month and, during a few dreadful months, to fifty and even thirty. 
At one of our Tuesday Partners’ meetings a little later, Fiasco came in 
with a plan. “We call it ‘The Mule without a Head,’’’ he said, 
passing copies of it around the table. 

We all looked at the document while Fiasco sat there with a 
sardonic grin we didn’t quite understand. First, there was a projec¬ 
tion for the coming year that showed the unit would make a small 
profit. No mean feat, considering the difficult times the elec¬ 
tronic scale plant faced. Then came a balance sheet that also 
forecast a recovery. Projected sales volume, however, was not 
optimistic, and investments in new products and production 
machinery was estimated at zero. So where was this profit to come 
from, we wondered? 

The answer was on the next page, where there was a list of all the 
people who would be left at the unit after the Mule without a Plead 
reorganization occurred. Absent were Rogerio, the purchasing man¬ 
ager, the sales manager, and several other key people. 

'How will you run the unit without all these people?” Vendramin 
asked, assuming Fiasco would be willing to put in extraordinary 
personal effort, out of self-preservation. 

“Do you see my name on that list, Joao?” Fiasco replied. 

That’s when we realized what the title of the proposal meant. 
There was an awkward silence around the table. 

“What will you do?” Batoni asked Fiasco after a while. “Do you 
have another job lined up?” 

Not at all, Fiasco said. I’m at the company’s disposal. I 
can t think of anywhere I would fit in now, but we can worry about 
that later. The bottom has dropped out of our business, and I can’t 

try to preserve my job when I feel the unit can’t afford me any 

Maverick a 215 

“This could all be temporary,” Clovis said. “What happens if 
the business springs back? It has in the past, you know. Are we 

going to throw away all the investment we have made in you and 
your staff?” 

“Today, Clovis, my recommendation would be yes,” Fiasco said 
solemnly. “If it were my money, I wouldn’t put any more of it into a 
business with such a risky future.” 

Does that mean you don t believe in the scale business?” 
Vendramin asked. Do you think we should shut it down?” 

“No, I believe in the business and in our product,” Fiasco said. 

What I don’t believe in is Brazil’s ability to pull itself out of its 
economic slump so soon. If you want my opinion, I’d move the unit 
back to Ipiranga.” 

Everyone protested at once. The decision to split up our plants had 
been one of our great successes. It was inconceivable that we should 
abandon this approach. 

The plant had made $1 million for us in its brief life. The trouble 
was, we had sunk half of it back into the building, the design 
consultants, the prototypes, and the specialized machinery. As we 
debated the future of the plant, Fiasco proved to us that we would 
have to spend more than $500,000 a year to keep the unit going as it 
was currently constituted until the economy rebounded. And God 
only knew when that would be. Cutting the payroll and folding the 
digital scale into Ipiranga would at least give us a chance to ride out 
the lean years without a heavy additional investment. 

So we moved it back, but only under certain conditions. First, we 
would keep the unit configured just as it had been at Jabaquara, 
except on a smaller scale. An area of 15,000 square feet was set aside 
at Ipiranga for the new plant, and everything about it—all invento¬ 
ries, decisions, operations, even the entrances—were to be separate. 
Of the twenty-eight employees at Jabaquara, twenty would move 
with the unit. Some of the other managers were transferred to other 
Semco units. The personnel reductions, incidentally, cost us another 
$1.5 million, since most people left with the equivalent of half a 
year’s pay or more. 

And what of the Head of the Mule? It would be a shame to lose 
someone who had had such a successful career at Semco and so 

216 A Ricardo Semler 

understood our culture. But we had nothing suitable for Fiasco and, 
given our opposition to stockpiling talent, putting him on the shelf 
was not an option. He stayed on at Semco for a few months, helping 
to organize the move and smoothing out the transition. Then, as 
destiny would have it, he became production director for Sasib, a 
subsidiary of the Italian giant Olivetti that made biscuit factories in 
competition with our Santo Amaro unit. 

“C'est la vie,” a Frenchman might say. Or perhaps, “M.erde” 



You can run a successful business or be ethical. Take your pick. 

You hear this a lot in Latin America, with good reason. Corruption 
wasn’t invented on that continent, but it is rivaled only by Africa as 
the warmest greenhouse for its growth. Below the equator it is a 
daily affair, democratic in the broadest sense, since everyone can take 
part if they wish. And so many do. Passports ordinarily take weeks to 
arrive, but can be had within a day if the way is greased. Building 
permits? Speeding tickets? Tax audits? No problem. 

How did 1 get to be such a nun at the brothel? I was born wealthy, 
which makes a huge difference. Having a family of means creates the 
opportunity for all sorts of idiosyncrasies. One of mine is proving 
that business can be conducted without blind obedience to estab¬ 
lished but anachronistic rules and traditions, including corruption. 

Most large Latin American companies wouldn’t survive the pres¬ 
sures exerted by crooked governments (which are the vast majority of 
governments in the last hundred years) had they not decided to trade 
commendable credos and cash for, say, a zoning change to build a 
factory or a license to open a branch business. But about three years 
after 1 took over, an opportunity presented itself to test my convic¬ 
tion that Semco could be different. 

Companies in Brazil are subject to routine inspections every so 
often, ostensibly to check compliance with building codes and other 
regulations. We received such a visit just after we renovated part of 

218 A Ricardo Semler 

the Santo Amaro plant. The law said we had to have a covered 
walkway separating the factory from the workers’ dressing area, but 
we didn’t. I don’t really know why. An oversight, I guess. The two 
inspectors decided on a small fine and gave us ninety days to comply 
with the requirement. 

But we were just too busy to get around to it. Alas, when payoffs 
are a way of doing business, people rarely feel the need to comply 
with arbitrary rules, since they believe all will be certified in the 
end, one way or another. 

The two inspectors came back and this time they said they would 
have to fine us $200,000 and, if we didn’t fix the building within 
thirty days, close us down. Pause. Or they could do some 
“consulting” for us, for $20,000 in “fees.” There would be no fine 
and no plant closing. Just business as usual. 

Clovis and I discussed the matter at length. Paying the inspectors 
the bribe was out of the question. But the possibility the plant 
would be shut was too awful to contemplate, and there was no way 
we could make the changes in thirty days. So we decided to blow the 

W^e knew the risks. Vfe would instantly become a favorite target 
for inspectors of all stripes. Retaliation was their forte. 

We called the mayor s office anyway and were referred to a high 
government official, who, we were assured, was an honest man. 

Off we went to see him. We had the names of the two inspectors 
and the last name of the man we believed was their leader, which 
was Rota. The official received us warmly and was extremely 
interested in the case. He was looking for opportunities to rid the 
city of corrupt officials, he said, and was thrilled we had stepped 
forward. I want the man who is leading my team of corruption 
busters to hear your story,” he said, picking up his telephone to 
summon him. 

Clovis and I exchanged a glance. “Hello,” the official said into the 
phone. Yes. Can you drop by my office right now? I have two 
gentlemen here who are telling me about a payoff.” . 

Minutes later a tall, white-haired man strode into the office. “Ah, 
here’s our man now,” the official said to us. “Mr. Semler and Mr! 
Bojikian, I’d like you to meet Mr. Rota.” 

Maverick a 219 

# # # 

Visits from inspectors were more frequent after that, but we 
weren’t sure Rota had anything to do with it. Most of the inspectors 
seemed honest, and we did all we could to stay within the rules. 

We weren’t especially worried when one day in 1989 a tax 
department official arrived at the Santo Amaro plant and began an 
audit. He was sixty years old or close to it, and more than a tad 
arrogant. Later, we learned he had been a professor and a lawyer. 
Usually tax inspectors stay about a week and then, if they are honest 
and everything is in order, move on. This guy spent a week with us 
and found nothing. Then he spent another week. And another. He 
was still at the plant after five months. Five months! Through it all 

our people scrupulously followed his instructions and gave him 
everything he wanted. 

Finally, the inspector had a meeting with our accountant and told 
him he had found “several irregularities.” If he reported them, he 
said, Semco would be liable for $700,000 in taxes and fines. Our 
accountant looked through his report and found it outrageous. All 
the inspector had come up with were a few administrative and 
clerical mistakes, all of them minor. An invoice with one date would 
be listed with another date in our books. Innocent errors, that’s all. 

'This is absurd,” said our accountant angrily. 

To his astonishment, the inspector agreed. “But if I levy these 
fines against you,” he added matter-of-factly, “it will cost you a 

fortune in lawyers’ fees to defend yourself. Sure you’ll win, but at 
what price?” 

So now we got the picture. For $150,000, he would save us all 
that trouble and a good piece of change, and give us a tax audit that 
would keep us safe from other inspectors for three years. They all 

worked together, the inspector said, so we were guaranteed not to be 

We had a quick meeting of top managers and unanimously 
decided to take another stab at whistle blowing. I phoned a friend 
who had been Brazil s finance minister and told him the story. He 
referred me to a high government official, who explained what we 
would have to do. Brazilian law is sticky about entrapment; only the 

220 A Ricardo Semler 

victims of a shakedown can conduct a sting. The police would have 
to wait until a payoff before moving in to make an arrest. 

Off Clovis went to buy the recording equipment. Oswaldo Guimaraes, 
the engineer, would set it up. Jose Violi Filho would conduct the 
final negotiations with the inspector and make the actual payoff. The 
plan was to hold another meeting at which he would tell the 
inspector that we accepted the deal and set the time and terms for 


The meeting was arranged for 3 p.m. that Friday. “I can laugh 
about it today, but I was really nervous then," Violi recalled. "The 
office where the meeting would be held was on the ground floor of 
our headquarters. The door would be closed, of course. But I was to 

turn the knob twice. That was the signal to Clovis and Oswaldo to 
start recording us." 

Violi must have seemed a little crazy that day, pacing up and down 
the hall and working his Fingers into tight Fists. But when the 
inspector arrived, he played his part perfectly. 

"I said it wasn’t going to be easy to pay him, since Semco didn’t 
have an ‘unofficial’ bank account for that purpose. He told me how to 
do it. I was acting like I didn t understand him, to make him repeat 
himself, because the more he spoke, the better it was for us. I said it 

would be really hard to get all that money at once. He said we could 
pay in ten installments.” 

The inspector even offered Violi a 10 percent cut, saying it was 
traditional that the person with whom he did business get a share. 

"I said I wouldn’t take it," Violi said. "That made him nervous. 
He only kept 10 percent, too, and the other 80 percent stayed in the 

system, which was the word he used for corruption. I had to say 1 
would think it over." 

Violi set up another meeting on the next Monday, same place and 
time, and promised he would have the First payment ready. The 
inspector said he would write up a report citing Semco with a minor 
infraction, which we could easily contest. "Don’t worry," he said. 
"I’ll show you how." But Violi, ever a detail man, insisted the 
inspector give us a clean bill of health. The inspector got nervous, 
but Violi was adamant. "Take it or leave it," he said. 

All this time we were next door, listening to bits of the 

Maverick a 221 

conversation. I say bits because the microphone wasn’t working 

properly and at one point a compressor went on in the factory and 
drowned out everything. 

By Monday we had perfected the recording system and arranged 
for the compressor to be silent. We also hid a video camera in a 
corner of the room, in some plants. At around 2 p.m. two police 
officers and two investigators arrived in an unmarked car. A little 
later, a reporter and a photographer arrived from one of Brazil’s 
biggest newspapers. We had invited them to document the arrest, 
just in case the government officials let us down. 

The payoff was set for 3 p.m. It was midsummer and unbearably 
hot in the un-air-conditioned building, especially in the bathroom, 
where six grown men were now hiding. 

We watched the clock: 3:00 came and went, then 3:10, 3:20, 
3:30. The inspector had always been prompt. Was he on to us? To 
help calm his nerves, Violi called home and chatted with his wife. It 
was 3:40, 3:50. The police officers, stretching, looked at their 
watches and wondered whether they should leave. 

Finally, the security guard rang to say the inspector had arrived. 
Everyone scampered into position. Adrenaline levels soared. The 

inspector apologized to Violi for his tardiness. His wife wasn’t well, 
he said. 

Violi got straight to the point. Are you sure you want to do 
this?” he asked, trying to make the inspector go through the scheme 
all over again, for the benefit of the recorders and, we hoped, the 
witnesses and eventually the courts. The inspector said he did and 
got out the audit papers for Violi to sign. Meanwhile, he told Violi 

how his wife had broken her leg in a fall over the weekend and was 
now hospitalized. 

Great, Violi thought to himself. In a moment he will be arrested 
and his life will be ruined. 

After they finished the paperwork, Violi turned over the check, 
announcing: This is the first payment.’’ 

“A check? I want cash.” 

Violi told him there was no way he could withdraw that amount of 
cash. The inspector wanted Violi to go with him to the bank. Violi 

222 A Ricardo Semler 

said he wouldn’t. Finally, the inspector took the check and put it in a 
little wallet. 

It was over. Violi invited the inspector to have coffee on the third 
floor. That was the signal for the police officers to arrest him. Violi 
hall expected them to break down the door and burst in with guns 
drawn. Instead, the door slowly opened and a single officer calmly 
entered the room. When the inspector realized what was happening, 
he turned pale, then looked over at Violi and started cursing. The 
police officers got the video cassette and the audio tape and sealed 
them in an evidence bag. As they all walked to the parking lot, Violi 
tried not to look at the inspector. The inspector asked to talk with 

his son, who was waiting in a car parked out front. Then everyone 
went to the police station. 

Violi, rattled by the episode, took some time off. 1 went on 
television to try to encourage other executives to help root out 

corruption. The inspector was eventually convicted of extortion and 
sentenced to two years in prison. 

There was no reason to rejoice, though. The inspector's superiors, 
whom he cited in the tapes, were not even brought in for question¬ 
ing. I got a few letters and telegrams of support, but when I 
proposed to establish a center to combat corruption and provide 
businesses with legal advice, not one executive publicly supported 

A little while later one of our clerks went to a government 

department for a document and was told, "Tell your boss he doesn’t 

just have one inspector to worry about, he’s got 100,000 against him 

now.” And so it seemed. Shipments destined for Semco plants were 

held up for ages at airports and docks. Semco trucks were stopped at 

state borders for inspection. Certificates, approvals, and forms of all 

kinds took forever to be issued. It was the price of doing business 

A year and a half later, while Sofia and 1 were on our honeymoon 
several inspectors showed up at the house we were building and 
demanded a $6,000 bribe. Otherwise, they said, they would hold up 
work for months. (They didn’t know who the owner was.) 

The contractor, long accustomed to paying, negotiated a sum. His 

Maverick a 223 

plan was to pay them before I returned, knowing I wouldn’t agree, 
then bill me many months later, when everything was done. But we 
returned before the payment was made. On our first visit to the site, 
I asked a foreman if everything was going well and he enthusiasti¬ 
cally agreed. “Especially since those inspectors stopped coming 
around,” he added with a grin. Alarm bells went off and I asked the 
contractor what had happened. 

I called the mayor, a woman known for her honesty, and told her 
the story. She asked me if I wanted to make another arrest. So we did 
it all over again. This time, Brazil s main television network covered 
it live. But I didn t take part in the proceedings or give interviews 
afterward. The mayor’s office handled everything and few viewers 
even knew it was my house. 

But, of course, the inspectors friends did. During the rest of the 
construction we were visited by an unending stream of inspectors 
armed with electronic measuring devices and thick books of building 
codes, most of which they probably hadn’t read in years. There was 
no end to petty bureaucratic obstacles. But it just hardened our will 
to build our home despite complex and unreasonable building codes 
that had undoubtedly been created to foster corrupt activities. In 
Brazil, we call this creating difficulties to sell simplicities.” 

Now we are the proud owners of a home that should be a museum, 
since it is probably one of the few structures in Brazil that meets 
every regulation. 



Next to our Santo Amato plant is a cemetery. Our silent partner, 
some of us call it. 

“See that lawn?” my father used to say. “It’s filled with people 
who were indispensable to their companies.” 

One day he too would be buried there, facing the factory at his 
insistence, “to keep an eye on it.” 

His special vigilance did not dissuade three of our most free- 
thinking engineers from making without doubt the weirdest proposal 
we ever entertained at Semco. They wanted to invite all the Partners 
to the cemetery, where everyone—dressed in black, naturally—would 
gather around a coffin emblazoned with a ribbon that read semco. It 
would be a symbolic burial of the company. 

While others complained that Semco was moving too fast, these 
three instigators—Oswaldo Guimaraes, Marco Aurelio, and Rogerio 
Ottolia (of the late, lamented Jabaquara unit)—were annoyed with 
what they regarded as our slow, cumbersome, wasteful, bureaucratic 
ways. Laying all that to rest was their way of waking us up. We 
didn t hold the ceremony, but engineers—who comprised an extraor¬ 
dinary team called the Nucleus of Technological Innovation—had 
once again made their point. 

The three had proposed the unit themselves at a meeting with 
Clovis, Laura, Marcio, and me. Their idea was to take a small group 
raised in Semco’s culture and familiar with its people and its 

226 a Ricardo Semler 

products—them, naturally—and set them free. Removed from day- 
to-day activities, they would no longer worry about production 
problems, billing, inventory, machines that didn’t work, or subordi¬ 
nates who wanted a raise. They would have all their time free to 
think. They believed that, thus liberated, they would invent new 
products, refine old ones, devise market strategies, unearth cost 

reductions and production efficiencies, even dream up new lines of 

It seemed to us that the three engineers had the right credentials. 

Each had shown a combination of creativity and pigheadedness that 

constantly caused them to question nearly everything about the 

The bearded, thirty-year-old Oswaldo had twinkling eyes that lit 
up every time the “Eureka!" machine in his mind switched on, 
which was often. He had started as a draftsman at the French 
machine tool company Brevet and rose to assistant manager before we 
whisked him away to become our chief engineer and then engineer¬ 
ing manager. In our Great Reorganization, he had become a Coordi¬ 
nator in charge of a sophisticated team that designed large machinery 
sue as mixing equipment for bubble gum and rocket fuel. He liked 

to say he turned his subconscious on when he went to sleep, and had 
the answers upon awaking. 

Marco, an electrical engineer, was a tall, awkward fellow with 

bulging eyes disheveled hair, and a strong contrarian personality 

that threw colleagues and customers off balance. A brilliant techni- 

cian, he would invent wiring schemes that no one but he understood. 

ine of biscuit machines he designed for United Biscuits had six 

miles of wiring, ,n fact, with hundreds of controls and thousands of 

switches. The company's engineers pronounced it “worthy of a 747 ” 

confiding to us that they would have settled for something on the 
order of, say, a 707. 

After Rogerio left the digital scale plant, courtesy of the Mule 

without a Head plan, he became our chief electronic engineer. Like 

Oswaldo he had often been courted by potential employers, but had 

spurned their lucrative offers. The NTI, it seemed, would make his 
loyalty worthwhile. 

We all liked their idea for the unit, but added five ground rules. 

Maverick a 227 

One, members wouldn’t have a boss. They would report to no one at 
all. Two—the flip side of one—they could not hire any subordinates. 
Three, they would be free to set their own schedules, write their own 
job descriptions, determine their own activities, and change any of it 
as they pleased, anytime and for any reason. Four, they would report 
their activities twice a year to the Partners, who would decide 
whether they would keep their jobs for another six months. Five, 
they would continue to receive a salary, though it would be less than 
they had been getting as senior managers. But they would also share 
in the proceeds of the ideas and innovations they thought up, 
whether it was profit sharing on products they redesigned, royalties 
on sales of new products they developed, or a percentage of the 
savings from cost reductions they came up with. (Flow big a cut 
would be up to them.) And they could sell consulting services to 
anyone else who wanted them. 

The three had been making between $25,000 and $35,000. As 
NTI members they would earn more than that in a good year, but 
less than they might as independent entrepreneurs. Then again, they 
would suffer less in a downturn than if they were on their own, and 
would have the backing of a large organization with a well-known 
name. All in all, we figured their compensation could swing from 
$15,000 to $80,000. 

They knew a good deal when they saw one. We all shook hands 
and our three newly liberated engineers went home to rest. The next 
day they would have to start thinking for a living. (Well, Oswaldo 
would anyway, since both Rogerio and Marco went off on vacation.) 

The trio moved into a small suite of offices in our Santo Amaro 
headquarters. Partitions that separated work areas were removed and 
drafting tables were installed. Soon, files started arriving from all 
over the company containing technical information on our products 
and processes. The engineers gathered all sorts of records and 
documents, probing for weak spots they might turn into opportuni¬ 
ties. Soon there was a large collection of spare parts—bolts, bits of 
scrap metal, an odd motor shaft, a gear for a dough mixer—piled in 
the center of their floor. To some it looked like junk. To the NTI it 
was inspiration. 

228 A Ricardo Semler 

Having thus feathered their nest, the team started to adjust to a 
life without routines. They arrived early, read the papers, and, well, 
thought. Then they thought some more. Like subjects in an isolation 
tank, they floated all day, unfettered by corporate gravity, undisturbed 
by bureaucratic distractions. 

It was too much of a good thing for Marco. He missed running 
the wiring of his machines, redesigning and then testing his ideas, 
and, yes, having subordinates to do those little things for him. And 

he was too nervous just sitting around thinking, so he went back to 
his old department. 

Laura de Barros, by then a Coordinator for training and 

organizational development, took his place. While Oswaldo and 

Rogerio concentrated on the industrial side, she would act as a 

counterweight on nonproduction issues, focusing on Semco’s 

That was what we kept telling ourselves, anyway. But at first the 
three of them butted in all over Semco. “Poking the wound,” they 
called it. The marketing department wasn’t aggressive enough, they 
complained. Why did human resources have to run the Semcotour 
program? Shouldn’t each department get to play host on a rotating 
as IS' And the y wanted the Partners to start a program in which we 
would call customers at random and ask them how satisfied they 
were with our products. 

We interfered with everyone, from Partners to workers,” Oswaldo 
recalled, barely suppressing a grin. “There are still people who don’t 

One afternoon, Oswaldo, Rogerio, and Laura were sitting around 

studying one of our mixers. It blended paint and pigments in a tank 

using a mixing arm that moved in a slow, circular motion. Laura I 

mechanical neophyte, asked how the tanks were cleaned when the 
colors were changed. 

"The arm is lifted and someone scrubs the tank,” Oswaldo replied 

Someone should invent a giant toothbrush to clean the tanks ” 
Laura said. 

So they did. The new product had a brushlike appendage that. 

Maverick a 229 

hooked on our mixer, swept the sides and bottom of the tank 
automatically, eliminating manual labor. 

Soon our curious trio of thinkers started attracting attention 
outside Semco, including an article in Exam, the big Brazilian 
business magazine. After that they were bombarded with inventions, 
including—this was my favorite—a device that looked like a minia¬ 
ture umbrella and opened up inside the nose, removing any 

They passed on the nasal cleaner, but even so had eighteen new 
projects under way by the time of their first six-month review. They 
grew even more prolific after that. In just a few years the NTI has 
come up with such diverse items as a scale that weighs freight trains 
while they are barreling along at full speed, a fiber optic tube used 
by doctors to examine patients' throats, and electronic systems to 
control small assembly lines. 

The team also formed an environmental consulting firm that 
advises clients on the recycling of waste materials and conducts 
audits to assess the environmental liabilities of companies being 
considered lor takeover. This eventually became an independent 
business unit, Semco Environmental Resources, and has, besides a 
permanent staff of tw r enty-five professionals, another forty biologists, 
geologists, hydraulic engineers, and economists on call. 

The NTI has made internal changes at Semco, too, reducing 
downtime on machines, streamlining manufacturing processes, and 
cutting assembly times. It has redesigned our paint mixer completely, 
changing the steel casing and the motors pulley system, which 
allowed us to cut its sale price by 32 percent and enter an export 
market that had been closed because of its high cost. And it has 
revamped our dough mixer, which used to take ten hours and 
fourteen steps to assemble but now requires three steps that take less 
than an hour. The gears, formerly custom-made, are now adapted 
from off-the-shelf automobile gears, another cost-saving idea of the 

We've even begun a junior version of the NTI. We call it “Lost in 
Space. Every year (provided it's a good one) we choose at least one 
young person from among our applicants from business or engineer¬ 
ing schools or even high schools. These extremely fortunate souls 

230 ▲ Ricardo Semler 

have no job description, no boss, no set responsibilities. They are free 
to roam through the company for a year, so long as they work in at 
least twelve departments and try to generate enough revenue to cover 
their salary. (Even a trainee can put together a financial analysis, 
gather marketing data, work in production, or sell.) At the end of 
the year, they are free to negotiate a more permanent arrangement 
with any of the departments in which they served. 

Weve unearthed some exceptional people through Lost in Space, 
including one young man who left us after two years to become, at 
age twenty-five, the planning manager for Shell Brazil. He went on 

to land a job in Paris as European sales manager for Brown's, which 
owns 4,000 clothing stores. 

We’ll let the multinationals train him some more, and then bring 
him back from space. 



Regimentation is the soul of the modern factory, and standardized 
shifts are the soul of regimentation. So what was that worker doing 
showing up at the darkened, deserted Ipiranga factory at 4:30 a.m.? 

Given the traffic, the long commutes, and the usually inadequate 
public transportation many workers put up with, it seems utterly 
unreasonable to expect them to time their trips so that they will 
arrive at the factory at precisely the same time, day after day. \£t 
every company does, even in huge cities such as Sao Paulo, with 
fifteen million people, most of whom seem to be on the road 

We had long ago let our office employees decide when they would 
start and end their work days, becoming one of perhaps a half dozen 
companies in Brazil offering this freedom. In 1988, we tried to 
extend it to our factory workers. If they wanted to have breakfast 
with their kids or take them to school, who were we to stop them? 
Beyond that, flexible working hours demonstrated our belief that we 
wanted to pay workers for results, not merely their time—and we 
didn t care how those results were obtained. Our idea was to agree on 
our common goal, then let our employees loose to achieve it. 

Anyway, if workers arrive promptly at 8 a.m. and leave at the 
stroke of 5 p.m., but are ineffective in between, what good is it? 
With flexible work schedules, we thought it would be possible to 

232 A Ricardo Semler 

finish the year with the performance we wanted but without mind- 
numbing records of what time people showed up each day. 

Such freedom was—is—unheard of in an industrial setting. At 
least we weren't aware of any other company, in Brazil or elsewhere, 
with factories running on flextime. When Paulo Pereira brought up 
the idea just about everyone at Semco was against it, including many 
of the plans beneficiaries. We had grown so quickly that as much as 
•rO percent of our work force had been at Semco for less than twelve 
months, and their experience elsewhere had convinced them that 
anything a company proposed was suspect, even flexible working 
hours. What s more, their union leaders, who like to make speeches 
about how employers ought to treat workers with dignity and 
respect, were jealous and insecure. When a Semco initiative appeared 
counter to common practice, they would drag their feet until they 
were sure they understood all the implications. 

So flextime had two strikes against it even before we considered 
the leery reaction of our managers. True, many of our products were 
being assembled by teams of workers in self-managing manufactur¬ 
ing cells, and workers were making many decisions bosses used to 
make. Even so, in many instances these workers all had to arrive at 
more or less the same time. Otherwise, production sequences would 
be scrambled and some would be idle, waiting for colleagues who 
were responsible for the previous manufacturing steps. 

For three months, Paulo conducted a feasibility study on flextime 
hoping to prove it would reduce absenteeism and poor performance’ 
One of his best sources was our "Good-bye Interview," in which 
everyone who was laid off, fired, or left voluntarily had a chance to 
give us a piece of his mind. "I live too far from work and my 
commute is too long," many would tell us. Or: "I haven’t had 
breakfast with my kids in ages.” Or: "I constantly fought with my 
boss about tardiness.” From all this Paulo concluded that starting 
times were a ma,or cause of conflict and misunderstanding between 
bosses and subordinates. He believed workers would let resentment 
over rigid schedules build, until they exploded at critical times. 

Paulo called our top managers together and told them that with 
flexible schedules it would be possible to reduce almost to zero 

Maverick a 233 

absenteeism and greatly cut overtime as well. “There are very few 
activities with such a high degree of dependency that it is not 
possible to have an interval of fifteen or thirty minutes between one 
production step and the next,” he told them. “In those cases, we 
wont adopt the flexible schedule. You’ll be pleased with the results. 
Conflicts with the supervisors will be reduced, the time clock will no 
longer be viewed as a monster, workers will feel more respected—and 
they’ll produce more.” 

No, they weren’t convinced. In response to their objections, Paulo 
said we would form a task force to mediate disputes between workers 
over starting times (which, in fact, has yet to meet). The managers 
still weren’t convinced, but at Paulo’s urging they decided to go 
along, if only as an experiment. 

The Nacoes Unidas marine products plant was Paulo’s laboratory. 
Though we had only opened in 1988, some of our most veteran 
workers, the ones who knew us the best, had been shifted there. We 
called in the factory committee members and the union leaders, who 
naturally brought their lawyers, since our plan would require major 
changes in our contract. 

Under the plan we had in mind, everyone would still work eight 
hours a day, but they could get to the plant anytime between 7 and 9 
a.m. and leave accordingly in the afternoon. 

Now, when our workers arrived at the factory most of them would 
go to the locker room, put on their uniforms, then go down to the 
cafeteria, have their coffee and rolls, and maybe read the newspaper, 
and then start their shift. They wouldn’t be clocked in, though, until 
they actually started working. We didn’t want to change this. But 
the union leaders’ attitude about flextime was: “No one just suddenly 
lets people come in whenever they want. We don’t know what it is, 
but there must be a catch.” And failing to find it, they took the 
opportunity to demand that workers be clocked in as soon as they 
arrived at the plant. 

We believed working hours were just that, and rejected the 
paternalistic notion that once you enter our gates you are under our 
wing.” We were pleased our workers felt at home at Semco and 
encouraged them to play cards, lounge about, read the paper, and all 

234 A Ricardo Semler 

the rest. But we weren’t going to change our policy of clocking them 
in when they arrived at their machines. 

After many meetings, neither side had budged. So the union did 
its best to convince the workers that, under the circumstances, 
flextime was a bad idea. With a truck and a loudspeaker, they 
repeated the well-known refrain: “Be careful, colleagues. Bosses don’t 
do anything good for workers.” 

But they didn’t prevail, perhaps because the employees knew us so 

well. During a mass assembly in the cafeteria the workers voted to 
give the plan a try. 

Most of them came in from 6:30 to 7:30, which was fine, since 
there wasn’t an assembly line at the plant. 

On to the Ipiranga factory. The union wasted no time mounting 
its attack there, too, putting out the word to workers that they had a 
right under the labor law to be five minutes late a day, or thirty 
minutes a week. It accused management of trying to eliminate this 
grace period. We argued that under the same law workers could be 
docked pay or even suspended for arriving more than five minutes 

late, and that, besides freedom, we were offering relief from that 

We successfully countered the union's attack only to face resistance 
from the managers, who wanted to scuttle the program before it 
started, believing that the workers were not capable of self-control. 
True, this had been a showcase plant, with interdepartmental teams 
and workers making all sorts of decisions. But don't forget that it 
had gone through the disruption of being divided into three units 

and workers had a set of new bosses to deal with. Plus, there were 
still scars from the strike. 

But the factory committee, under the leadership of Soares, saved 
us. First, it convinced the workers to give the plan a try. Then it 
stationed a group of workers near the time clock to discourage 
colleagues from violating the new rules. At the end of the first 
month the guilty parties were identified publicly and warned that if 
they didn't comply with the program they wouldn't have to worry 
about what time they came in, since they wouldn't have a job 
It was at Ipiranga that Saulo Henri, a welder, decided to 
come in at 4:30 a.m. But one worker alone in the dark could get 

Maverick a 235 

hurt, so we asked him to readjust his hours, and he complied. 
Eventually, nearly half the shop-floor workers decided to come in at 
the same time, 6:30, so that the dishwasher assembly wouldn’t be 
disrupted. Many commuted along the same rail line, and there was a 
train that was nearly empty at that hour. 

Flextime was more of a problem at the machinery plant at Santo 
Amaro, where obstreperous and irresponsible union leaders persuaded 
workers to vote the plan down. But a year later, the members of the 
factory committee, impressed with the success of flextime at other 
plants, called a plantwide meeting to reconsider the idea. Flextime 
was adopted by a nine-to-one margin. 

Alas, it was a different story at our air conditioner plant. The town 
of Diadema, where it is located, has a large concentration of metal 
workers, a contentious bunch who are represented by an avowedly 
Marxist union with links to the Worker’s Party, a political organiza¬ 
tion that comprises many far-left and militant groups. 

Workers at this plant, and members of the factory committee, 
were much more radical and less trusting than anywhere else at 
Semco. Years later they eventually came around, and their leader, 
Vicentinho da Silva, told a television audience that “Brazil had only 
one trustworthy boss—Ricardo Semler and Semco.” 

But back then they didnt want to try flexible hours. In fact, it 
wasn’t until 1991 that they finally accepted the idea. 

In high school I was fascinated with the high jump, maybe because no 
one else was. It was as if I was always measuring myself against the bar. 

erched atop two supports, it seemed to enjoy thwarting my efforts to 
surmount it. 

In the beginning, five feet was a grand obstacle. Observing the 
intensity of my interest, the coach decided to teach me the then-radical 
technique of jumping backward. After almost a year of practice, I was 
clearing 5'8". Every inch became an obsession. The bar and I had long , 
solitary matches. Sometimes I would take the school bus home frustrated 
beyond words. Sometimes it would be the bar s turn to brood. Making it 
over 6'0" was the most important goal in my life. I would barely scrape 
the bar and lie breathless on the mat, peering up to see if it would roll 
over onto me. Then after one jump it held fast. A silly grin formed on my 
face and stayed there for days. 

A few years ago, I struggled with an opportunity to acquire a company 
with five plants and 2,000 employees. “Why do we want to grow more?” 

I asked myself. Are we going to be better for it?” It reminded me of a 

feeling from an earlier day. Why do I constantly throw myself at the bar? 
Is there no limit?” 

It s all about persistence, isn t it? But where does persistence end and 
obsession begin? How high is too high? How big is too big? Of course, 
some growth is necessary for any business to keep up with competitors and 
provide new opportunities for its people. But so often it is power and 
greed and plain stubbornness that make bigger automatically seem better. 




In the first eleven years I led Semco, Brazil had two good years, 
three transitional years, and six dreadful years. Inflation, the scourge 
of Latin America, averaged more than 400 percent a year, swinging 
from yearly highs of 1,600 percent to lows of a mere 100 percent 
(with some months of deflation in between). From 1986 through 
1990, the country endured five economic shock plans, knocked three 
zeroes off its currency twice and on two occasions changed it 
altogether. The stock market soared during the country’s worst 
recession and went bust during a boom; banks made billions when 
inflation rose, then fired 150,000 clerical workers when it fell. 

Given this almost unimaginably complex and difficult economy, 
150 million people were nearly always unsure what to do. A 10 
percent wage hike might look tempting one month, then prove 
inadequate when the official statistics would later show that inflation 
had been 15 percent. The tension between suppliers and customers 
was intense, as each sought to change the payment terms to their 
own advantage. Competition to sell anything was fierce, but Semco 
endured the dangerous oscillations with the help of our workers. W^e 
had years in which sales grew by 80 percent and even 120 percent, in 
real terms, but also years in which they dropped by 20 percent and 
then 35 percent the next year. 

As the 1980s the lost decade,” the experts called it—ended, 
the economy tell off a cliff. Industrial output plummeted by 9 

240 a Ricardo Semler 

percent in 1990 and the gross national product was heading back to 
levels not seen since the 1970s. In Sao Paulo, there were half a 
million people unemployed, a number that would nearly triple by 

And just when we thought it couldn’t get worse, it did. A new 
president, Fernando Collor de Mello, assumed office and appointed a 
young economist, Zelia Cardoso de Mello (no relation to the presi¬ 
dent), as finance minister. She proceeded to test some new theories, 
including one that held that there was too much money in circula¬ 
tion, that it belonged to too few people, and that they were doing 
too much speculating with it. Because of this, her theory went, not 
enough money was being invested in industry. This was generating 
inflation and stagnation. 

So, she thought, let’s take some of that money and give it to the 
government (which doesn’t have enough, right?). On a sunny spring 
day in 1990 she went on television to declare a bank holiday and 
seize 80 percent of the cash in the country. The government laid hold 
of savings accounts, checking accounts, certificates of deposit, com¬ 
pany funds, the works. Every Brazilian, no matter what his assets, 
was left with $800 or 20 percent of his holdings, whichever was less. 
If someone had, say, $1,000 in a checking account, he now could 
spend $200. The lady said she’d give the money back, corrected for 
inflation by an official index, in twelve monthly installments, 
starting in a year and a half. 

Chaos doesn t begin to describe the reaction. Industrial output 
sank another 14 percent in the next twelve months. Companies 
didn’t have money to meet their payrolls, much less to conduct 
business. At Semco we struggled through several months of zero 
sales what company was going to buy machinery that took ten 
months to deliver when it didn’t know if it would survive the 
week?—and then the government decided to open the country to 
imports. The duty for foreign machinery, which had been 45 
percent, was cut to 35 percent, then 30 percent, and by the end of 
1992, to 20 percent. So we had to contend not only with a market 
that had lost nearly half its volume but also with increasing competi¬ 
tion from abroad. 

Thrashing about in precarious waters, we started discussions with 

Maverick a 241 

our factory committees to find ways to keep Semco afloat. It struck 
most of us as obvious that we could no longer employ the same 
number of people as we had. Not with so few sales. The question 
was, how could we all work together to minimize the pain? 

Nothing is more corrosive to motivation and productivity than 
layoffs, and over the years we had tried mightily to insulate our 
workers from the wrenching volatility of the Brazilian economy For 
starters, we refrained from hiring people to work on products we 
knew would have a short life span. Opening a Hoola-Hoop division 
without knowing what to do with all those Hoola-Hoop makers 
when the fad inevitably wanes might make you rich in the short run, 

but it is a strategy for adventurers—and adventurers have no future,' 
only a present. 

Also, we never promised anyone job security. In fact, we refrained 
from hiring good executives if they demanded a job contract. There 
is no such thing as a sure thing, and one side or the other will 
necessarily suffer. Look at the damage those golden parachutes have 
done to corporate morale, not to mention balance sheets. 

Nearly everyone stops putting people on the payroll when business 
is down, and almost everyone eases up when it improves. This is 
another error. Sales start going up and soon secretaries, office 
assistants, chauffeurs, analysts, receptionists, and an army of others 
are streaming through the doors. Sales slow, and they are all 
“reevaluated,” depressing the lucky ones who survive the inevitable 
purge. Even when the government stimulated the economy and 
orders abounded, Semco controlled hiring with the same iron hand it 
used during lean times. We possibly lost market share as a result, 
but when demand sank to its original level we didn’t have to let 
scores of the new people go. Any alley cat can stay lean when food is 
scarce; the trick is to stay lean during the good times. 

Yet from 1989 to 1991 Semco suffered a 40 percent drop in sales. 
We had come to rely on our workplace initiatives for at least partial 
immunity against economic turbulence, but there’s only so much you 
and your people can do when you lose almost half your business. For 
some of our plants, there was no way out. 

242 A Ricardo Semler 

In the late 1980s, our marine unit at Santo Amaro enjoyed a huge 
$14 million backlog of orders. Since it only produced $5 million 
worth of products a year, it had three years of work already secure. 
We decided to invest more than $500,000 in the building, which 
had an innovative layout the workers helped design and new pump¬ 
testing facilities that cost a bundle. It all seemed worth it, though, 
as productivity ascended to unimagined heights. 

Then Hurricane Zelia hit and the shipbuilding industry took the 

blow head on. Emaq, the shipyard that had saved us in 1981 with 

the Petrobras order, foundered. Then Verolme, a yard that had 

ordered pumps for five ships from us, followed. Only three shipyards 

were left, and they were months behind in paying their bills. Semco 

was left with $1.5 million of receivables that were no longer 

collectible and, even worse, $4 million worth of products ready for 

delivery that the shipyards could no longer pay for. We needed 

20,000 square feet just to store all the unwanted pumps, half-ready 

casings, and 300 horsepower motors. The stacks were fifteen feet 

We called the workers together and discussed what we all could 
do. One proposal, designed to avoid layoffs, called for salary cuts of 
20 percent across the board, until the business revived. But many 
workers, already struggling with their bills, felt they couldn’t afford 
even a small cut. They also thought it would be like “using a sieve to 
block out the sun.” These employees wanted us to take our losses and 

lay off part of the workforce at once, and they had enough votes to 
turn our proposal down. 

We tried all the cost-slashing measures we could think of—coffee 
breaks were cut to once a day, copying machines were locked away, 
electricity consumption was monitored, new uniform purchases were 
suspended, and all expenses were looked at by many sets of eagle 
eyes. But 1 didn't hold out much hope. I'm not a big proponent of 
cost-cutting programs. 1 like to think we don't spend money 
unnecessarily even in good times. And how do you measure how 
many sales were lost because the reps had their gasoline allowance 
cut, or what the cash flow might have been if the clerks in billing 
had not been saddled with a cut in telephone expenses, or even how 
many little mistakes and miscalculations could have been avoided if 

Maverick a 243 

engineers hadn’t cut back on photocopies of blueprints? When the 
foolish penny pinching is over, everyone goes back to business as 
usual—until someone thinks expenses are excessive again and begins 
a new round of cuts, starting with expenditures that were needed to 
make up for the shortsightedness of the previous cost-cutting 

Our people at Santo Amaro scrimped and saved, but it was far 
from enough. So we got serious. We organized the factory workers 
into teams and sent them out to sell replacement parts at the docks. 
This was a steady market, since incoming ships needed to change 
worn impellers, shafts, and wear-rings, boom or bust. But lots of 
smaller machine shops had already realized this. We called these 
shops pirates because they made cheap, shoddy copies of Semco parts 
and gave “friends” on the ship fat commissions to buy them. Our 
workers did their best to compete, but they were no match for all 
these Captain Hooks. 

Meanwhile, at the marine products plant, a worker assembly was 
called at the cafeteria and we repeated our proposal to cut working 
hours and pay to keep everyone employed. This time, we said, we 
would implement the wage reduction in a Robin Hood—like manner, 
so those with lower salaries wouldn’t suffer as much. 

Labor law has always been paternalistic in Brazil. (Justifiably so, 
given the greed of some Brazilian businessmen.) If a worker is paid, 
say, $250 a week, his company has to pay him another $1,000 at 
Christmas, $1,300 at vacation time (four weeks paid, plus a $300 
bonus), and deposit a sum equal to 8 percent of his wages each 
month in a savings account. If an employee is dismissed, he gets 
whatever is in that account, plus 40 percent more, plus double pay 
for vacation days that haven’t been taken. Then there is severance pay 
that, for employees such as ours, many of whom have been with 
Semco ten, fifteen, even thirty years, would add up to two years of 

It wasn’t a golden parachute but, given the belief of many workers 
at the marine products unit that our business was not likely to revive 
soon, it wasn’t a bad deal, either. Why work for one or two years at 
reduced wages, only to be laid off anyway? So like the workers at 

244 A Ricardo Semler 

Santo Amaro they voted down our salary reduction plan, in the 
process consigning some of their colleagues to walk the plank. 

About this time I went to Ipiranga to see Alipio Camargo, then its 
Partner. Sales of our big industrial dishwashers had dropped from 
forty a month to twenty-five and then to a measly five, and inventory 
levels and factory expenses were killing us. If the situation didn’t 
improve, I didn t have to tell him, the plant would have to shut. 

I was so absorbed in my thoughts as I left that day that I didn’t 
notice Joao Soares, the factory committee leader, standing in front of 
the welding booth. But he saw me. 

Joao knew things were bleak, but he was adamant. The plant, 
which had been proof of what can be achieved when workers are 
treated as adults and encouraged to make decisions, could not be 
shut. Not now. Not after what we had accomplished. 

Tell me, Joao,” I said, “if this business were yours, and you knew 
that over the next quarter you were going to sell only a handful of 

dishwashers a month, not the forty you used to sell, what would you 

“That’s the trouble with you,” he replied, “you’re only thinking 
about the next three months. What about when we start to grow 
again? Or do you believe we’re never going to grow again?” 

No, Joao, it s not that. But it’s been a while since this unit has 
made any money.” 

“There are things we can do,” he said. 

The machines were down. In the background I could hear the 
murmur of workers as they said their good-byes and left for home. 

“I’ll leave it to you,” I told Soares after a while, “and the factory 
committee. Come up with a plan, go over it with Alipio and the 
others, and we’ll try it. But think it over, Joao. If you have the 
capacity to produce forty dishwashers a month but you sell only a 
handful, what are you going to do with all those people? The 
economy could take years to recover.” 

It s so much easier to throw the stone, I thought to myself, than 
be the window. 

Joao went to his house just outside Sao Paulo—a house he could 
not have imagined owning when, as a nine-year-old, he sold iced tea 

Maverick a 245 

and peanuts at the Maracana soccer stadium in Rio. He stayed up 
long into the night worrying about the plant. He thought about it 
all through the next day, a Saturday, and Sunday, too. Semco was his 
Too much so his constant presence at the plant was one reason 
why, a year earlier, his marriage had failed. 

Joao thought about the succession of grim factories in which he 
had worked before joining us. One had toilets with half doors, so 
workers couldn’t hide from foremen. At another, he would lose a 
whole day’s pay for being one minute late. Pacing up and down his 
small bedroom, he thought about how different his life now was. He 
remembered his first factory committee meeting. He and his co- 
workers were too scared to talk. And the benefits they had acquired! 
health insurance, free breakfasts, flexible working hours, and, most 
important, the intoxicating feeling of self-determination that is all 
but nonexistent in conventional factories. The workers at the food 
service equipment unit had achieved so much, Joao thought. Then it 
struck him. That was their way out. 

When he arrived at Ipiranga on Monday he gathered all 150 
workers in the middle of the shop floor, telling them he had an idea 
that might save the plant and their jobs. The workers would 
voluntarily reduce their wages by 30 percent and forgo a 10 percent 
raise they were due. They would also give up subsidized meals, their 
transportation allowance, and other benefits. That was the conven¬ 
tional part of the plan. Then came the twist: the workers would take 
over all the services at the plant provided by outside contractors and 
third parties and perform them themselves, slashing the company’s 
costs. At the cafeteria, Soares said, the workers would buy the food 
and prepare it. They would guard the factory gates and clean the 
offices and the shop when the workday was over. They would 
transport finished goods to customers and sell replacement parts to 
restaurants and hotels. 

In return, the workers would share with management the author¬ 
ity to run the plant, making all business decisions jointly—a 
guarantee that their sacrifice would not be wasted. All strategies, 
policies, and investments—every check—would require approval 
from both the bosses and the workers. Oh yes, and the bosses would 
consent to a 40 percent pay cut. 

246 A Ricardo Semler 

When he finished explaining his plan, his colleagues were silent. 
Then a woman spoke. 

Clean the bathrooms? No way. I’ve never done it, and I’m not 
going to start now.” 

Ill be the first to clean the bathrooms,” Joao said. ‘‘There’s 
nothing wrong with spending a half hour a day cleaning bathrooms. 
And the offices. Look, with this plan the company won’t have to send 
anybody away. We won’t have any layoffs.” 

WLen he called for a vote, more than a hundred hands were raised 
in approval. 

The plant managers, as might be expected, were hardly as 
enthusiastic about Soares s idea, but slowly they warmed to it. Alipio 
was already thinking about starting his own business and would soon 
implement a Mule without a Head plan of his own, essentially 
putting himself out of a job for the second time in just a few years. 

Some changes were negotiated. Management accepted the 40 
percent pay cut for two months only, with a 30 percent reduction 
after that. The cuts were applied in the Robin Hood fashion, to 
protect lower-paid workers, and—ever the optimists—employees 
were granted an extra profit-sharing payment of 13 percent, on top 
of the 23 percent provided by SemcoPar. And with that, Semco had 
its first experiment in comanagement. 

After just one month, we could hardly believe the results. The 
workers had saved so much that the extra profit-sharing clause was 
actually invoked, which helped compensate for reduced salaries. 

The second month was even better. And by the end of the third 
month, the employees’ salaries had been fully restored. Sales of 
dishwashers stabilized at a hardly spectacular twelve a month, but 
the plant was selling more spare parts than before. 

Even so, comanagement, though a noble experiment that helped 
stave off a deep round of layoffs, wasn’t a permanent way to run the 
Hobart plant, or any plant. Profitability depended on an extraordi¬ 
nary low level of expenses, which in turn depended on exceptional 
efforts of workers that couldn’t be sustained forever. 

And there are other reasons why 1 don’t believe in comanagement. 
Take that second signature on the checks, the one that belongs to the 

Maverick a 247 

workers. Do they get to appoint anyone to sign? What if they choose 
someone to lead them who won’t work with management? Joao 
Soares was particularly well suited to deal with the problems we had 
at the food service equipment unit, but what if the workers there 
decided they wanted, say, a radical union leader or someone who 
wasn’t versed in our corporate culture? 

Comanagement requires two sides, which is one more than neces¬ 
sary to run anything efficiently. Too many leaders may be worse than 
too few. We had proved that at Semco as we eliminated extra levels of 
management. This was no time to turn back. 

All in all, we withstood the downturn better than other companies 
in our fields, most of which laid off 30 percent to 40 percent of their 
workers. We had endured a 40 percent drop in sales and hadn’t even 
increased our borrowing. But carrying many surplus workers on the 
payroll and finishing new orders without the inflow of cash from old 
ones had cost us nearly all of our retained earnings. 

After months of debate, we concluded that Brazil’s economy 
wasn’t going to get better soon, so continuing along this path was 
dangerous. We didn’t want to change our philosophy, or even our 

long-term strategy, but we needed some new tactics if we were going 
to survive. 

Well, we had created a company as flexible as any we knew. If 
nothing else, Semco had been redesigned to adapt to change, and to 
do so quickly and without preconceived, “this-is-the-way-it’s-done- 
here solutions. Now we would put that adaptability to the test. 



Pedro Miranda de Oliveira paced up and down, wearing out the 
linoleum in the hall outside the cafeteria at Santo Amaro and 
demonstrating why he was called The Bullet. He was facing the 
biggest decision of his life, and he had only an hour to decide. 

a short, muscular man with dark, tousled hair and eyes 
that darted all over the place, had joined Semco in 1979 as a lathe 
operator, running the big machines that cut metal bars into shafts for 
pumps, mixers, and motors, it required considerable skill, since a 
wrong move would turn dozens of pounds of valuable bronze or 
stainless steel into scrap. Pedro became one of our best operators, 
then went through a nine-month training program studying machin¬ 
ing and learning to read engineers’ plans. After that he moved up to 
our most demanding lathe, which could cut cylinders of steel five 
meters long into shafts for agitators used for refining gold, among 
other things. 

In 1987, Pedro became a Coordinator, supervising more than fifty 
co-workers who cut and polished sheet metal for mixers and pumps. 
We weren’t surprised when he proved good at it, for he made the 
quick decisions blue-collar workers prize. 

But now, despite our efforts to keep the payroll tight, our success 
in cutting unnecessary expenses, and our astonishing gains in pro¬ 
ductivity, Semco was in trouble. We had too many employees 
making too many products at too many factories—or maybe just too 

250 a Ricardo Semler 

few customers. Either way, we had to change. We had achieved a true 
partnership with our workers—er, I mean our Associates—based on 
trust and a mutuality of interests. But in this excruciatingly harsh 
economic climate, what we needed was a divorce—an amicable divorce, 
of course, with community property. 

During our Chats at Lunch sessions, at assemblies with workers, 
in interviews company officials gave to magazines and newspapers, 
Pedro had heard the phrase horizontalization.’’ We were constantly 
discussing and debating what functions Semco should continue to 
perform and what activities it should farm out to others. As we 
thought more about it, we became convinced we no longer wanted to 
do anything that could be done just as well elsewhere. 

But at Semco there is always a wrinkle, isn’t there? Instead of 
contracting out business to strangers, we decided we wanted to 
contract it out to the people we knew best: our workers. We would 
help them start their own companies, transforming themselves from 
employees to Partners. And so the Satellite Program was born. 

If you want something done well and cheaply, do it yourself." 
That was the mantra of the bigger-is-better companies. But how 
many businesses have lost their way as they grew? Henry Ford was so 
fond of verticalization he raised trees to make the sideboards of his 
Model Ts, bought iron mines and cargo ships, even searched the 
Amazon for a site for a rubber plant for tires. The company’s official 
history doesn’t play it up, but Ford had to fire 60,000 workers 
because of this rampant do-it-yourselfism. 

Farming out work to specialists, subcontractors, consultants, and 
assorted third parties can avoid a host of problems. For starters, it 
reduces fixed labor costs. It also helps empty the shelves of inventory, 
since the raw materials and spare parts a company would normally 
stock art spread out among its new suppliers. Companies like ours, 
which make products requiring complicated and sophisticated manu¬ 
facturing processes, tie up a lot of capital in stockrooms and 
warehouses. This is not good when the economy takes a downturn. 

Above all, people act differently when they own their own 
businesses. Workers who fight for every extra minute of a coffee 
break will toil late into the night and on Saturdays and Sundays if it 

Maverick a 251 

means keeping their own company alive. At Semco, we had 
succeeded largely because we had increased our employees’ stake in 
their jobs. Our people already worked late and on the weekends of 
course, and they didn t need any prompting from bosses. But by 
encouraging them to start their own businesses, we would raise their 
sense of involvement even higher. 

Theoretically, farming out work means a loss of the profits of 
verticalization. But just ask Henry Ford about that. So as Brazil’s 
economy foundered, we decided to heed lessons old Henry ignored. 
Few companies could match our five-meter lathe, a rare machine that 
can cut down huge metal cylinders, or our vertical mills, which are 
used for creating perfectly polished and symmetrical surfaces on steel 
casings and housings. Few companies had the expertise to solder the 
huge internal coils of our cooling towers, fabricate special gears for 
our mixers, or coat couplings with titanium. These operations would 
be expensive if not impossible to subcontract, so we would continue 
to perform them. We would also keep our staff of highly specialized 
design engineers, who adopt our products to suit customers’ specific 
requirements, whether they needed mixers for soup or bubble gum. 
Same with the people responsible for computer-aided design work 
that not only entailed expensive equipment but also crucial trade 

But drafting in ink? There were thousands of drafters in Brazil as 
skilled as ours. And sheet metal work? There were hundreds, if not 
thousands, of Brazilian companies that could roll steel plates for 
mixer bodies. And what about legal work and software development? 
Weren’t these candidates for subcontracting? 

Subcontracting is not without problems. Third parties need to 
learn a lot about a client company’s business, and that isn’t easy at an 
idiosyncratic company like ours. There’s also the “brain-drain” factor, 
and the risk of letting outsiders know your know-how. But in our 
case, both worries were unfounded; under the Satellite Program, we 
would be working with people we already knew and trusted, and vice 

But how would we convince employees by the hundreds to become 
entrepreneurs, leaving a secure nest at Semco in the midst of an 
economic storm? That’s where those paternalistic Brazilian severance 

252 A Ricardo Semler 

provisions came in. The six-, ten-, or even twenty-plus months of 
salary we were obliged to pay employees when they left, plus 
additional benefits Semco added over the years, would be their seed 
money. Of course, no employee would have access to his nest egg 
unless they were fired. So we would offer to fire them, then help 
them use their severance to establish their own businesses, which 
would supply materials or services to us. To clinch the deal, we 
offered to lease our workers the very machines they operated in our 
plants, at no cost to start with and extremely reasonable terms later 
on, as their companies became profitable. 

Once they set up their own shops, our workers would have the 
possibility of making many times what they could earn at Semco if 
the economy straightened out. Yes, that was a big IF. And if the 

recession persisted they might make less than they would at Semco_ 

but only assuming they continued a job at Semco, which was, for a 

distressingly large number of people, becoming more doubtful with 
every day. 

Pedro wasn’t surprised when he and the other Coordinators at 
Santo Amaro were called in for a “serious conversation” with top 
management. He knew business was bad. 

He had already been through one layoff, going over the numbers 
with executives, helping draw up a list of colleagues to be dismissed, 
discussing the choices with the factory committee, and then breaking 

the news to the employees, one by one—leaving time in between to 
recover his composure. 

Pedro was expecting to go through it all over again now. He knew 
about the Satellite Program, but he thought it was one of those 
things that only happened to other people. So he was taken aback at 
the proposition we made. Out of the dozen employees who had 
endured through the hard times and were still under Pedro’s supervi¬ 
sion, we suggested that half would remain at Semco. They performed 
sophisticated operations that were part of our technological core. We 
hoped many of the rest of the workers would be employed at a 
company that might be called Pedro and Friends, Ltd. 

Pedro was so stunned he uncharacteristically asked for time to 
think it over. After an hour or so, he stopped pacing and returned to 

Maverick a 253 

the boardroom, where the managers were waiting. As he entered, he 
felt, he said later, “like a bride-to-be.” 

‘All right,” he said, taking a deep breath. “I’ll do it.” 

The actual name of his company turned out to be JBL Machining 
and Sheet Metal Ltd. The J and the L were for Jose Maria and Jose 
Lima, two Semco colleagues. (The B was for Bullet.) The three of 
them spent several long days searching before finding a suitable 
building a few blocks from the plant. They drew up the list of 
machines, all of which we delivered to their new address, and, after 
squeezing them into the small shop, Pedro and his colleagues began 
supplying us with couplings, gearboxes, drives, metal shafts, turbine 
blades for mixers, and machine casings for marine pumps. 

Their lives changed drastically. At Semco, Pedro was always home 
by 5:30 p.m.; now his wife sometimes takes him coffee to his little 
shop at night, and when Pedro finally does get home he often brings 
paperwork with him. Semco set up a team of executives to teach our 
mini-entrepreneurs to control costs, set prices, manage inventory and 
maintenance, and take care of all those bureaucratic details. There 
was no end to their problems, as we knew, since we did business 
with them. “I would send them the same invoice twice,” he recalled, 
“and they would call asking whether 1 had doubled my price without 
telling them. Then I would hear laughing on the other end of the 

But we never tired of helping him and the others. And we were 
only too happy to call their other prospective customers and provide 

Like other Satellite firms, JBL has had good and bad times. Its 
payroll has soared to ten employees—God knows where they put 
them—only to sink back to five. But they haven’t lost their 
enthusiasm or their optimism. In fact, they’ve just started a profit- 
sharing plan. 

No one was forced to start a Satellite company; all our employees 
could get severance with no strings. Some took the money and 
simply left. Others tried to hang on to the payroll as long as they 

But our offer was almost irresistible. 1 know companies that tell 

254 a Ricardo Semler 

suppliers that they cant sell to competitors, or that dictate the price 
or profit margin. We told our people they were free to sell whatever 
they made to whomever they wanted, even our competitors. Then 
again, we would be free to buy from any company we wanted. There 
were no guarantees for either side, which insured that we would be 
both competitive and innovative. 

The Satellite Program spread quickly from plant to plant and 
office to office. White-collar employees actually took it up first, 
especially our tax people, human resources staffers, and draftsmen. 
We dissolved our legal department and farmed out the work to 
several firms with different specialties, including one formed by one 
of our ex-lawyers. Some of our accountants formed a firm, too. And 
computer programmers went off on their own to make our software. 

Starting with Pedro and his colleagues at Santo Amaro, the 
Satellite movement caught on with blue-collar people at the food 
service equipment and refrigeration systems plants. I won’t say it was 
utterly painless. Let’s say we had a business unit with five manufac¬ 
turing cells, and four detached from the mother ship and went into 
their own orbit as satellites. It didn’t make any sense for us to keep 
the remaining cell; we’d just be competing with the other four. 

There were some surprising converts, as well. Tired of pollution 
and traffic, Paulo Pereira had long wanted to move his family from 
Sao Paulo to his native Bebedouro, a small town 150 miles away. But 
we didn’t want to lose Paulo, who was the architect of some of our 
most innovative programs. The Satellite Program was his chance and 
ours: Paulo set up his own consulting and recruitment firm in 

Bebedouro, and commuted to Sao Paulo to work for us Tuesdays 
through Thursdays. 

Laura de Barros went off on her own, too, to become a human 
resources consultant. Simpliciano Domingos de la Sierra hung on, 
although at one point the young entrepreneur had to drive a minibus 
in order to support his family. And Alipio Camargo, with a handful 
of Semco colleagues and a pair of mechanics, started a company to 
sell and service Hobart products in Sao Paulo. He’s doing well. 

Semco has so far helped form more than two dozen Satellite 
companies. About half the manufacturing we had performed in house 
has been turned over to them, and we believe we can farm out 

Maverick a 255 

another 10 percent or even 20 percent in the coming years. To this 
day, no Satellite has closed. Some are looking for partners, others 
struggling to expand product lines. Some are little Semcos, organ¬ 
ized around the ideals of democracy, transparency, and trust. Some 
are utterly traditional, tiny Fords or IBMs. (A temporary affliction, 
we hope.) Almost all have customers besides Semco, which is fine 
with us. It’s their business. 

The Satellite Program works because it is based on the principle 
that people who have a stake in their company are bound to be more 
involved in their work. As a result, only good things will happen: 
costs will fall, quality will rise, innovation will bloom. People will 
look at a part and say, Why does it have to be like this? Why can’t it 
be made better? Or cheaper? Or faster? 

For Semco, the Satellite Program has meant tremendous flexibility. 
We buy only what we need, when we need it. Freed from the 
distractions of manufacturing, we can concentrate on designing, 
engineering, and assembling better products. We no longer have all 
those expensive machines sitting around that we feel compelled to 
use, so we’re not locked into procedures and processes. 

In a more fundamental sense, the Satellite Program is an extension 
of our philosophy of empowerment. After all, our new entrepreneurs 
have complete control over their workplace—at least the control any 
owner has. They make all the decisions, including deciding how 
many decisions their workers should make. They are almost always 
even more productive than they were at Semco. 

Implementing the Satellite Program wasn’t easy or quick. But I 
think it has helped us remake Semco into a company that can float on 
the roughest seas, without taking on water or having to force the 
crew to abandon ship. 



Suddenly, the plants where we all had worked so hard to work 
together were half empty, the machines and the people who ran them 
scattered all over. We looked around and asked ourselves, “Do we 
need all these factories anymore?” It was time to move out. 

None of the plants were easy to close, but the most difficult was 
the food service equipment unit, where in the years since we took it 
over billings and sales had more than quadrupled, even as manage¬ 
ment was pared in half. We lost close to $1 million, a horrendous 
sum, closing the plant and moving what was left to Santo Amaro, 
which made the decision that much less comprehensible to some 
workers who had been based there. “What company buys a plant, 
spends money to redo it, introduces programs that improve produc¬ 
tivity and turn the place around, and then walks away from it?” we 
were asked more than once. But moving the food service equipment 
operation would cut expenses to a level they couldn’t match if they 
stayed. Rent on the old plant alone amounted to 6 percent of sales, 
which was about three times the usual percentage. 

The BAC unit in Diadema, which made refrigeration equipment 
for the food and beverage industries and cooling towers for air 
conditioners, was also hard to close, since its business was holding up 
much better than our other plants. Its main customers, beer and soft 
drink manufacturers, were relatively unscathed by recession; they 

258 A Ricardo Semler 

weren’t subject to much competition from abroad, and their custom¬ 
ers, though pinched, still had change for a soda or a beer. 

But as the Satellite Program caught on at the refrigeration systems 
unit we saw that there was the potential to farm out 60 percent to 70 
percent of the plant’s production, without a drop in quality. The 
move to Santo Amaro was inevitable, although we took it as a great 
compliment that leaders of the union there, a radical organization if 
ever there was one, tried to talk us out of leaving. The region was 
home to Ford, General Motors, and other companies with tens of 
thousands of workers, so the union was hardly worried about the 
hundred jobs we were transferring. But it had been using Semco and 
its programs as examples in its negotiations with other companies. 

Clovis and others felt closing the plants was a step in the wrong 

direction. Rationally, he could see the Satellite Program was our best 

strategy, but his heart was with those odd factories with colored walls 

and shrubs between the machines and workers who hardly needed 

The decision wasn't as wrenching for me, because I was further 
from the day-to-day management of the company. Unlike the 
workers, I didn't have to help cart off the furniture and the fixtures. 
That s the difference being a general who never sees the front line. 
Although it can lead to disasters like Gallipoli, often officers who 
aren t close to the shooting make better strategic decisions. 

Truth is, I had been thinking about closing the plants and 
consolidating our operations at Santo Amaro for several years. The 
idea was initially dismissed by just about everyone—Vendramin, 
Violi, Batoni, and Clovis. Even as the economy sank and our fate 
became apparent, it took months to bring everyone around. 1 know 
many people feel we at Semco talk too much before making deci¬ 
sions. They assume a company our size should turn on a dime. I 
admit it: we can take longer to make a decision than General Motors 
which is 10,000 times bigger. But if we debate a decision forever’ 
once we make up our minds we usually implement it much faster,’ 
since everyone is totally committed to it. 

Which is what happened when we closed our plants. Even Clovis 
came to agree it was the best way to preserve our idiosyncratic 
culture. We wanted a company built to last decades, one that could 

Maverick a 259 

withstand the ups and downs of many business cycles. We would 
never achieve that stability if we had to maintain a certain sales 
volume or depend on desperate, last-minute cost cutting to cover the 
fixed expenses of a large work force and machine-filled factories. 

Semco went rrom 830 employees and nine business units at five 
locations in 1987 to a bit under 300 employees in six business units 
at two sites (plus some 200 workers in satellite companies and 
full-time consultants) four years later. We rebuilt and expanded our 
original plant in Santo Amaro to house our divisions, each in its own 
area, and used part of the Nacoes Unidas plant to test marine 
equipment and as a warehouse. 

But the consolidation of our plants didn t afFect their structure one 
bit. We’re still a decentralized company. We’ve retained the essence 
of these units their autonomy, their individualism, their separate 
factory committees, their interorganizational teams, their personali¬ 
ties. In a sense, our divisions aren’t really in the same place at all. 
We ve divvied up the space so there is no interaction between them. 
Each unit has its own turf, entrance, storehouse, and shipping dock. 
Only their addresses have been changed to protect the transplanted. 

The divisions can negotiate among themselves to change the 
layout anytime they want. They are even free to move away, if it 
makes business sense. Recently, a small start-up unit called Difitex, 
one of the two or three embryonic operations we usually have in our 
orbit, concluded it was being charged too much for its space at Santo 
Amaro. (Each unit pays a share of the building’s costs, including 
security, cleaning, electricity, water, and insurance, based on its 
square footage.) So the leaders of the unit, which imports textile 
machinery, decided to move somewhere in downtown Sao Paulo. 
They didn t tell me, and I still don’t know where. It doesn’t matter. I 
send a fax when I need to talk to them. They fax me back. 

Although we’ve shrunk Semco to a core of sales, engineering, 
design, materials handling, purchasing, and assembly people, our 
plant looks as bustling as ever. Santo Amaro is always overrun by 
former employees who sell software, tax advice, audits, ball bearings, 
welded components, and slicers and fan scales that we sell under our 
name. There are hundreds of people wandering around, many of 

260 A Ricardo Semler 

whom I've worked with for years and a few I don't know from Adam. 

They can use our desks and phones and computers and park their cars 
in our lot. 

When some of the consultants whose advice we seek from time to 
time on strategy, organizational behavior, and marketing heard we 
were going to consolidate, they advised us to pare down our product 
lines at the same time. Retain only those goods and services that 
made money, they said, such as marine parts and pumps, large 
dishwashers, food and beverage cooling equipment, specialized bis¬ 
cuit machines, high-tech mixers, and our environmental engineering 
and auditing team. Eliminate weaker products such as Hobart 
sheers, fan scales, potato peelers, and meat grinders, and oil filters 
and air-conditioning equipment. But we decided these were still 
good products; we just weren't making them at competitive prices. 
Which is where the Satellite companies, with their inherently lower 
overhead come in. We take their products—sometimes finished and 
ready to be sold, sometimes parts in need of final assembly—tack on 

our brand name, and use our sales force to market them, at a 
comfortable margin. 

Consolidation has also enhanced our ability to innovate. We still 
perform the final assembly of most of our products, and that is when 
you sense the problems and opportunities in a production process 
And we still control the critical interaction between engineering and 
assembly. We can change from sheet metal to ceramics or from 
stainless steel to plastic anytime, and no machines will become 
obsolete because of a switch. 

Similarly, for twenty years we had made a pigment mixer for paint 

manufacturers that consisted of a large steel plate with teeth to do 

t e blending. Our competitors switched to mixers made of ceramic 

material, which didn't rust and were much cheaper, but we didn't 

consider a switch because we owned the machine that made the steel 

blender. Once we leased it to a Satellite company, we were free to 

(Cynics might accuse us of palming off obsolete equipment on our 
new suppliers. But most of it isn't. And don't forget, they were 
getting the machines at bargain prices.) 

The consolidation cost Semco more than $2 million. That, cou- 

Maverick a 261 

pled with the loss of more than $4 million from canceled shipyard 
orders, should have been enough to throw us into a lions cage of 
bankers. Brazil’s economic crisis brought down lots of fine 
companies—an average of 800 went broke every month. Semco not 
only survived, but also paid the huge cost of plant closures and 
employee reductions and readaptation without taking on more debt. 
We managed to break even in the very worst years and made good 
money in middling ones. Our employees, who each produced an 
average of $10,800 worth of goods a year in 1980, now produce 
$92,000 worth of goods a year (adjusted for inflation), four times the 
national average. And by the value-added standard, productivity rose 
six and a half times. Sales volume grew from $4 million a year to 
$20 million or so a year, with one third the workers. Yes, we’re down 
from a peak of $35 million in 1987 (with 830 employees), but it’s 
still a hell of a growth rate for such a miserable economy. Indeed, by 
the end of 1992 we had six months of working capital in the till, 
without a single outstanding bank loan. (Violi had a list of nearly a 
dozen bank executives who were seeking an appointment to try to 
talk us into correcting that situation.) We were prepaying suppliers 
and had increased our people’s real salaries—that is, above 
inflation by more than 7 percent. Moreover, almost all our Coordi¬ 
nators received fat bonuses in 1992, and we had begun launching 
two new businesses, a factory maintenance service and a waste 
disposal and recycling operation for offices, including container 
systems on every floor that separate glass, paper, and plastic and then 
automatically recycle it. 

We should have seen it coming, you say. 

We did. Some of us did, anyway. 

Three years before Brazil s economy collapsed and Semco launched 
its Satellite Program and then consolidated, we held a weekend 
retreat for forty of our top managers. It was a typical Laura de Barros 
production, the lights were low and Debussy’s La Aier was playing. 
Everyone was asked to lie on the floor, relax, and disconnect from 
day-to-day concerns. I was surprised how even hard-boiled engineers 
got into the spirit of it, kicking off their shoes and stretching out in 
all directions. I should have known not to underestimate Laura. 

262 a Ricardo Semler 

The key exercise involved "visioning” the future of the company. 
Each participant was asked to imagine what Semco would be like in 
the year 2010. What would the plant look like? How many people 
would work there? You get the idea. 

The managers pondered and then, when the lights came up wrote 

down their individual predictions. Then they were asked to share 

them with a colleague and together meld them into a single portrait. 

a mg around the room, I overheard a snippet here and there. 

see myself going to work in a train that has fax machines and 

computer terminals,” one man said, as his partner grimaced. 

y office is full of greenery—it looks like a rain forest ” said 

The shop floor is polished like a ballroom,” someone else says 
and the assembly line workers are all ,n bright white overalls ” ’ 

Then each pair of managers joined another pair and repeated the 
•ce ding process. Then those four met with another four, and so on 

was .. CraZy ' deaS had been scre ened out and what remained 
was a collective, integrated vision of what our top managers thought 
Semco would be like. mougnt 

What was the vision? 

eveTYf) XPeCted th3t ’ b , eC3USe ° f ° Ur annUa ‘ gTOWth of 40 P er «nt 0^ 
c. ,r en : ln the mid ' 1980s ’ our managers would picture 
Semco with perhaps 15,000 employees by 2010. But the company 

they described was not much larger than the Semco of the present^ 
a oug t e quality of both our products and the lives of our 
employees was much higher. Our people did not want a bigger 
company, they wanted a better company. A company in which people 
could work at home, liberated from conventional structured 
'chedules. A company in which there was such fluid movement onto 
and off of the payroll that it wasn’t always clear who was an employee 
and who wasnt. They didn’t want our businesses to cause pollution- 

:"Li:r»:i p ' ,n,s ,o * f “ -»- 

Somehow this vision of a smaller, more fluid, more flexible less 
ined company has mostly come to pass. No more are we victims of 

the f “p~p'e. planes, P ,r,:! 

Maverick a 263 

more revenue. We have outgrown the allure of growth, albeit after 
paying the price in money, time, and gastritis. 

To want to grow just to be big is an idea that comes from the 
sandbox. Sure, some growth is necessary for nearly every business. It 
allows for diversification of products and markets, which is one of the 
best ways for a firm to guarantee its survival. It creates additional 
opportunities for employees and improves motivation and productiv¬ 
ity, since it creates change all over a company. 

But beware. Growth opportunities are always springing up and 
should be regarded the way Ulysses regarded mermaids. Much about 
growth is really about ego and greed, not business strategy. At 
Semco we initially pursued the acquisition of companies because a 
good part of our own potential had been fulfilled. We studied more 
than a hundred firms, negotiated with fifteen, and bought four. I can 
summarize in three sentences the hundreds of hours and millions of 
dollars we invested: 

Growth through acquisition is exciting, glamorous,' and 

The company you buy is not very similar to the one you thought 
you were buying, and never like what they told you. 

Buying small, family firms is a certain way to skip the ulcers 
and go straight to bypass surgery. 

In our case, we incorporated subsidiaries of multinationals, which 
mostly honor commitments, God bless them. They usually have 
accurate books, unlike family-owned firms, where the closets are 
typically full of skeletons. But when you buy any company, you must 
be willing to watch it and learn from it, at least for a year, before 
putting your paws in the soup. 

Most of our acquisitions eventually worked out. But now we've 
consciously made the decision to stop growing. Our people want to 
be convinced the products they make are necessary and want to like 
making them. They want to end their careers with a feeling of 

The vision our managers shared with us was surprising at first, but 
it gave us the confidence we needed to consolidate. Simply grossing 

264 a Ricardo Semler 

$40, $50, $100, or even $200 million a year means nothing. If 
you re willing to take risks, it s easier to pile up revenue than build 
the kind of organization our managers told us they wanted. Today we 
are still expanding. Remember those 1,400 resumes I told you we 
had received when we placed an ad for engineering Associates? We re 
expecting to sign contracts with more than 150 of them in the 
coming months. But most won’t be full-time employees, they’ll be 
our partners. Similarly, we will grow our sales force through our new 
network of Satellite companies, consultants, and partnerships. 

1 ve seen countless companies grow tremendously and then flame 

out like a meteor. Those once-in-a-lifetime propositions occur all the 

time, don’t they? Whenever I’m tempted by a deal, I remember what 

Ray Krinker of Price Waterhouse used to say: “A small hole can sink 
a big ship.” 



Most companies, even conventionally pyramidal ones, practice at 
least some form of consultative democracy, which means key execu¬ 
tives can speak their minds, if only to one another, before the CEO 

This is better than Stalinism, since he didn’t listen at all. But it 
falls well short of democracy, in which all employees—not just 
managers—have a say in corporate decisions. 

Why do so many companies believe it is necessary or even 
desirable for one person to have the last word? I suppose they think 
its more efficient. But centralized power is a high-risk proposition. 
Henry Ford may be known as the man who pioneered the manufac¬ 
ture of automobiles (even though other companies were making 
them, on a smaller scale), but he was also a hardheaded dictator who 
had to dismiss thousands of workers because he mistakenly insisted 
on continuing the Model T well after its market evaporated. Alfred 
Sloan of General Motors may be regarded as one of industry’s most 
able organizers, but he was also a shortsighted structuralist who 
created a company so inflexible it failed to react to the threat of 
economy cars from Japan. 

Ford and Sloan, like Stalin, knew how good it feels to wield 
absolute power. But is it good for a company and its em¬ 
ployees to have a single strongman do pretty much as he pleases, 
with a disproportionate amount of attention to his habits and 

266 A Ricardo Semler 

with the instability that reigns when his successor needs to be 

Nothing is harder work than democracy, I keep telling myself 
I don't remember the last time I made a corporate decision 

f ° ne ’. "° r , Can 1 count a11 the times I've been voted down. But 
gladly bite my lip when I disagree with a judgment made 
y consensus, because I believe that unfettered democracy is much 
more important (and even more profitable in the long run) than 
prevailing over our managers in a way that takes you back to the 

ays in which seesaws and sandboxes were important parts of the 

an added benefit to having a democratically minded No 
1. the No. 2s, No. 3s, No. 4s, and No. 5s can play meaningful 
roes right away. Too many vice presidents in traditional com¬ 
panies are made to feel like also-rans; when they can't move up 

they have to leave. An inordinate amount of talent is lost this 

With this in mind, and with Semco restructured in a way 

that made it much less vulnerable to the economy, I decided it 

was time to virtually eliminate another level of our hierarchy 
mine. y 

Instead of one person at the top, Semco would be run by a 
committee of our Counselors. They were, I believed, a particularly 
-balanced team, professionally and personally: Clovis, a father 
igure who had been influential in my thinking; Vendramin, an 
economist, industrial manager, engineer, and, above all, a thoughtful 
man who took his time about everything; Batoni, who pushed hard 
for results and was much less forgiving about people than either 
C ov IS or vendramin; Viol, with a first-class financial mind and two 
on the ground; and Jose Alignani, a talented engineer and 
natural leader. They ranged age from the, fortief to th £ 

They now also own 1 percent of the company each and share in 
t e year-end dividends. This wasn't just L Lpres.on of gra^ 

in our tor"^ t0 foStCr , th l entre P reneurial spirit of capitalism 
.n our top management. I chose 1 percent because it is a large 

nough share of the company to be meaningful to each of the 

Maverick a 267 

executives, and because their joint holding of 5 percent is also an 
important block. 

They use their Tuesday meeting to discuss and make major 
operating and strategic decisions—that is, those that haven't been 
made by the Associates or Coordinators or resolved at the Monday 
meetings at each unit. Sometimes there is precious little for the 
Counselors to do. Other times, it seems as if everyone else has 
decided to stop deciding. 

Every six months, one Counselor takes a turn as acting chief 
executive, coordinating the activities of the Partners and their 
business units, representing the company in legal matters, and 
sometimes meeting with customers who insist on speaking to the top 
dog. This rotation avoids the excessively collective thinking that 
characterizes government agencies and gives each Counselor a chance 
to put his mark on the company (although the time any one is in 
office is short, so that mark won’t be indelible). 

So now I m just another Counselor. But my job hasn’t changed—I 
try to make things happen, like a catalyst. I lobby for what I believe 
in. I step in when I think I can do some good, and step out when 
I’m tired of an issue or when the other Counselors are tired of me. I 
attend their Tuesday meetings only when I’m invited, which is about 
every two or three weeks. Otherwise, they’re on their own. I’d 
probably be asked to more meetings if I didn’t have a habit of 
throwing out wild ideas. Also, I tend to have a hands-on approach 
that disrupts the system. WLen I become obstreperous, they just 
screen me off from the rest of the company. Like Mafiosi, none of 
them will talk to me about new developments unless he is accompa¬ 
nied by one of the others. 

It s just as well. At the risk of sounding immodest, I’ve been 
concerned that I’ll exert an undue influence on Semco. It’s important 
to discredit the belief that the company will survive only as long as 
I’m there. 

What if they make a decision I don’t like? I’d say at least 20 
percent of the time I disagree with them. Maybe 30 percent. This 
includes some major decisions in the last year or two. In 1991, for 
example, our books showed us in the black in spite of what had 
been, according to Brazil’s biggest business magazine, the worst year 

268 A Ricardo Semler 

ever for machinery manufacturers. Just before our balance sheet was 
to be released to the newspapers, Violi called to say that an 
adjustment in a customer account was going to turn the nominal 
profit into a $90,000 loss. 

“It seems so silly to report a loss when it’s such a ridiculously 
small number,” I told Violi. 

“But the profit would be small, too,” he countered. 

I know, I know, I said, feeling a bit exasperated. “But they’re 
different categories. A loss is a loss, and a profit is a profit.” As 

I said it, I realized that this sounded like “a mother is always a 

The Tuesday meeting was the next day, and the annual report had 

to be ready on the day after. Clovis asked whether I wanted to 

participate in the meeting, at which a final decision would be 

Were you going to invite me anyway, or is it because of the 
balance sheet?” I asked. 

"Well, we wouldn't have called you for anything else,” he 

“Then have the meeting without me. I'll fax you all a memo from 
home giving my opinion.” 

In my memo, I noted that there was a series of adjustments 
concerning our shipyard accounts as well as an import payment that 
our auditors said we could include either in this year's or next year’s 
budget. I didn't see why we shouldn't use one of these perfectly legal 
adjustments now to offset the small loss and get back our profit. 
After all, a profit in such a bleak year would look better to the 
vultures in the business community who were constantly circling 
overhead, waiting for Semco to slip up. I reminded my fellow 
Partners of the time we had laid off sixty workers at the marine 
products plant, and were accused in the newspapers of firing ”1 500 
employees in one bloody afternoon.” No amount of press releases 
could undo the damage that caused. So, I told my colleagues, let's 
not leave ourselves open to mischief for so little. 

I sent the fax off, confident my arguments would prevail. They 
held their meeting and decided in favor of reporting the $90,000 
oss. They had already told our employees about it, they reasoned, 

Maverick a 269 

and to report anything else risked looking as if they were cooking the 

Violi called to tell me the news. He said the Partners would 

meet with me to explain the decision. I asked him whether my 

arguments had been aired, and he assured me they had. That 

made me feel better. And when the Partners asked me to write 

the commentaries on the balance sheet for the annual report, I 

readily agreed, explaining why we had decided to show the small 

There were no hard feelings. It was just another day at a 
democratic office. 

Besides the money and the perks, I’m sure most CEOs enjoy the 
sport of running a company—devising strategies and products, 
crunching numbers, taking risks, screwing around with other peo¬ 
ples lives. It’s like being a five-star general. 

If I wanted to go to work every day at 7:00 and leave at midnight, 
I’m sure I could keep busy, as I did before my detour to the Lahey 
Clinic. But while most CEOs insist that they enjoy 70 percent or 80 
percent of their jobs, I suspect a more accurate percentage is about 
30 percent. There are so many meetings, phone calls, boring 
luncheons, administrative headaches. 

That s about how much of my time I spend on Semco now, 30 
percent. I like to think it’s the gratifying 30 percent, since I hardly 
ever do anything I don’t enjoy. I have no more than two or three 
business lunches a year, never leave the office (at the plant or at 
home) after 6 p.m., and return fewer than five of the twenty to 
thirty phone calls I get each day. 

I respond to the rest and most of my mail with handwritten notes 
that I fax from my home machine. When I first started working at 
home a half day a week, I imagined everyone thought it was an 
excuse for the boss s son to spend the day by the pool. I did nothing 
to discourage this suspicion, letting people who came by my house to 
pick up a report find me in shorts. But the flow of documents from 
my house soon made it apparent I was, in fact, working at home, 
and rather efficiently. Today I work there from three to five mornings 
a week. 

270 A Ricardo Semler 

Sometimes, just by sticking my nose into some issue, I’ll find 
myself saddled with half a dozen phone calls, faxes, and memos. So 
unless I’m the only one who can handle a problem, I’ll steer clear. I 
hate it when I make myself needed and then can’t pull away. I’m 
proud to say I no longer know what a Semco check looks like. I 
haven’t signed one for almost eight years. 

The truth is, the company hardly needs me in its day-to-day 
operations anymore. And the ideas Semco is built on aren’t mine, 
either. They flow from the company’s culture, and that belongs to 
everyone at Semco. I don’t have anything against capitalism, despite 
what my critics say. I value my shares in Semco, But it’s really not 
my company anymore. I am not Semco. Semco is Semco. 

I have even gone so far as to insure that this separation will extend 
to the next generation. I teel the company is too precious to run the 
risk that one of my children or grandchildren will badly manage it. 
And anyway, I don t believe in the family business in the long term. 
Family companies have a harder time attracting talented and ambi¬ 
tious professionals, who realize that at the very least the criteria for 
promotion won’t be straightforward. Indeed, anyone who has the 
time to calculate the percentage of family businesses that have 
survived for four or five generations will soon be encouraging his son 
or daughter to go to medical school. So my children will not have an 
assured place at Semco. I know what you’re thinking. All business 
owners say this. But I mean it. I’ve already seen to it that none of my 

offspring can be promoted without the approval of three quarters of 
the Semco board. 

Not even my death will change their circumstances, since out¬ 
standing shares in Semco will revert to the foundation that Irene 
Tubertim helped start, which will be managed by another twenty- 
one-member board that will include employees and outsiders but not 
family members. It is through such mechanisms that Semco will 

finally have its own personality, completely independent of any 
Semlers. J 

When I think about the traditional CEO, I often think about my 
father. And when I think about my father, I recall a line from a 
James Taylor song: "The secret of life is to enjoy the passing of 

Maverick a 271 

time.” Most people live either in their memories of the past or their 
hopes for the future. Few live in the present. 

I always wanted my father to enjoy the money he made. He never 

did. He was always worrying. Late in life, after his cancer had 

been diagnosed, he would walk through the park after radiation 

therapy and tell my mother that he had never really noticed the 

flowers and the ducks before. It took seventy-three years and a 

terminal illness to make him see the small but fascinating details 
of life. 

When I was a teenager, my father would take the family on a long 
European holiday at the end of each year. For him, it was half skiing, 
half business. I remember we would always have dinner in the 
hotel restaurant. There was a sign on the wall on which room 
numbers would flash, indicating a phone call. Our room number 
appeared constantly. When it was 8 p.m. in Europe it was 4:00 in 
the afternoon in Brazil, and people at Semco would be calling 
him, one after the other, to discuss that day’s crisis. They were 
delegating up, because he made it clear he wanted to know 

everything. No matter how far he was from the office, he couldn’t 
leave it. 

He had a stroke, caused by cancer in the liver, while on a ship off 
the Italian coast in 1985. It was the first time he had been on a long 

cruise. He thought he was finally removed enough from the business 
to be out of touch for a while. 

Looking back, I should have moved away from Semco earlier. But 
to its owner, a company is like an adolescent child. You want it to 
grow up and face the world on its own, but you lie awake worrying 
about whether it is going to smash up the car. 

I admit I still worry a bit when I am away for long periods. So 
much can change so quickly in Brazil, usually not for the better. But 
my fears have always been unfounded, and I work hard at suppressing 

them. Persistence is a virtue only when it is pointed in the right 

What do I do with the other 70 percent of my time, when I’m not 
working at Semco? I write a weekly newspaper column every Sunday, 
for 1.1 million readers. I talk about Semco to companies and 
business groups around the world. I am interested in politics and am 

272 A Ricardo Semler 

a member of the executive committee of a large political party, the 
Partido da Social Democracia Brasileira. I watch a minimum of three 
movies a week, buy the recordings of Beniamino Gigli, Billie 
Holiday, Philip Glass, and Shostakovich, and take piano, golf, 
Chinese, and cooking lessons. I also read fifty books a year, mostly 
histories of war and empires. Centuries of blunders and successes are 
there for guidance. For a handful of dollars you can buy an explana¬ 
tion that would have spared Napoleon thousands of lives. For little 
more than the price of a Big Mac you can discover the mistakes of 
the Tokugawa shoguns in leaving their camp at Nara unguarded. For 
less than it costs to fill your tank with gas you can find out why 
Winston Church ill, hero of World War II, couldn’t manage shortly 
thereafter to get himself elected. I also read four newspapers a 
day—they’re even cheaper. But I don’t spend much time on the front 
page. The causes of great wars, early signals of stock market 
collapses, and the advent of innovative technologies are not found on 

the front page, not at first anyway. Often the most important news is 

Then I have this list of goals. I chanced upon a television program 
about fifteen years ago that, as people so like to say, changed my life. 
It was an interview with a sixty-one-year-old American named John 
Goddard. I never found out who he was, or heard his name since. 
Anyway, Goddard had made a list of more than one hundred goals 
and set out to attain them. He only had nineteen to go. They 
included landing on the moon and living to the year 2000 He had 
already piloted a plane at the speed of sound, driven a submarine, 
descended the Nile in a canoe, navigated the length of the Congo 
River, and climbed Chile’s Aconcagua Peak. 

I decided that what worked for him would work for me, if on a 
less ambitious scale. I picked sixteen goals, about half of which I’ve 
already met. I believe I have turned Semco into the most sought-after 
workplace I know. I have created a foundation to provide opportuni¬ 
ties for poor Brazilians. I am five-sixths of the way toward speaking 
six languages fluently. 

And like Goddard I travel incessantly, and always with eyes open. 

I shudder when 1 think of those awful trips in the early 1980s, when 
Harro and 1 were desperately trying to keep Semco afloat. Now Sofia 

Maverick a 273 

an I set out for X.’an to see 6,000 terra-cotta soldiers buried 
hundreds of years ago, take balloon safaris in Kenya, camp in 
Tanzania, scuba dive in the Seychelles, cross the Sahara, comb 
beaches in Thailand, and float down the Nile. 

It'S a far cry from three cities in a day, then back on the red-eye. 

nd when I m off on one of my jaunts, I don’t leave a number where 
1 can be reached. And I don’t call in. 




I had just finished speaking to a group in the south of Brazil when 
a man in a white suit and white shoes approached the microphone for 
the last question. "This story of yours is all very interesting," he 
said, but I ve been sitting here for two hours waiting for something 
that I could use at my hospital and pharmacies, and I don’t know 
that I’ve heard anything that’s applicable.” 

Twelve hundred pairs of eyes turned toward me, and I could see 
that some people who were already on their way out had stopped. 
What could I tell this good doctor so that he wouldn’t have to go 
home disappointed? 

Let s take one of the pharmacies as an example,” I began. 

But they re much too small, he interrupted. "Two or three 
employees, no more.” 

That s big enough, 1 said. "Who determines where the contain¬ 
ers of medicine are placed on the shelves?” 

"My partner.” 


“They’re arranged alphabetically, of course. So they’re easy to 

I m not so sure, I continued, "but indulge me with a few more 
answers, and then I’ll make a proposal.” 

Everyone settled back to see how it would end. 

"How are the employees’ schedules made up?” 

276 A Ricardo Semler 

Our personnel department does that. It’s very simple: a rotation 
schedule, so someone is always in on Saturdays, and during the night 
and on Sundays and holidays when it’s our turn to be open,’’ the 
doctor explained. 

“These people are paid fixed salaries, right?’’ 

That s the way it s always been done,’’ he answered, a bit warily. 
Well, here’s the deal. You’ll give me the phone number and I’ll 
talk to your employees. And you’ll give me leeway to provoke their 
imagination. He had a skeptical expression. "Then you will give 
them time to decide how to run the pharmacy as they think best, 
and tell me what they did.” 

There were groans as people in the audience realized there wasn’t 
going to be a quick end to our little drama. 

“And,” I added, "we will tell all the people who are here today of 
the results. I turned to a reporter I had recognized in the front row. 
"Would you agree to write a story about it?” 

She nodded. 

‘Now, we only need the doctor’s consent.” 

111 take you up on it, he said. "Let’s do it with our smallest 
pharmacy. Its sales are less than 1 percent of our business.” 

“So I have your trust for that 1 percent,” I said, as the audience 

We met afterward and I learned a bit about the pharmacy 
business. Then 1 called his employees and told them about Semco. I 
spoke to them a few more times in the next week and sent them a 
copy of my book. The doctor, true to his word, gave them the 

freedom to try anything they desired. They talked to me about a few 
ideas, but then I lost touch. 

About six months later the doctor phoned me with his report. 
There were three employees in the pharmacy: two women with 
shopkeeping experience and a young man who was the pharmacist. 
All were single and disliked the irregular working hours, which put 
a damper on their social lives. So they started their experiment by 
asking that scheduling be transferred to them. They each worked the 

same number of hours, but devised a rotation that made it easier to 
plan their leisure time. 

Soon, they had rearranged the boxes of medicine. The alphabetical 

Maverick a 277 

arrangement, .t turned out, often required them to climb tall 
adders. In the new system, boxes were stacked according to how 
often the pills in them were requested. That meant aspirins were 
next to vitamins and heartburn tablets on a low shelf. 

Then they gave each product a number and bought an inexpensive 

computer to keep track of stock. They would take inventory at night 

and on holidays, when few customers came in, and reorder supplies 

t emselves, in the quantities they felt necessary, with a goal toward 

maximizing profits. They also suggested that the pharmacy stock 

more products, including cotton swabs, bandages, sunscreen, 

andruff-fighting shampoo, and other items not normally carried by 

hospital drug stores, which usually stick to the basics. In the 

beginning, the employees were a bit pushy with customers and even 

scared some away. But eventually they found an appropriate level of 

Last I heard, the employees had proposed a profit-sharing plan, 
and the doctor was spreading the Semco gospel at his hospital. 

Of course, the doctor, like me, owned the business. I get many 
questions from people in the middle or even at the bottom of their 
companies or organizations. They can’t just decree that things will be 
different. So, they want to know, what can they do to change things? 

I found myself in the middle not too long ago, when I was in¬ 
vited to join Brazil’s imperious Federation of Industries of the State 
of Sao Paulo. And I discovered that, with perseverance and courage, 

Semco-style management can flourish in the most hidebound of 

The federation counts among its members the nation’s most 
powerful industrialists, traditionalists all. It is unwise for a Brazilian 
president to appoint a finance minister without consulting this 
sixty-year-old club, and positions such as labor minister and the 
presidencies of the Central Bank, the Bank of Brazil, and the 

National Bank for Social and Economic Development are rarely filled 
without its benediction. 

As Semco became more prominent, I had begun writing articles in 
Brazil’s largest conservative newspaper, with 400,000 readers. The 

278 a Ricardo Semler 

paper is a mainstay of the elite, so it seemed an ideal place for me to 
rail against them and their federation. 

I imagined what the old guard must have been thinking, sitting 
in their leather club chairs and smoking cheroots. “Such a small 
company, you know. “He inherited it from his father, didn’t he?” “He’ll 

grow up to be like the rest of us one day. It’s the only way to stay in business 
in this country.’’ 

But at first my columns provoked the worst of all responses, which 
was no response at all. 

So 1 was truly startled when, having never ever set foot in the 
federations imposing, sixteen-story, pyramid-shaped (what else?) 
building, I was asked to become an officer of this regal group. The 
invitation came from Dr. Jose Mindlin, a friend and one of the few 
federation directors who believed that it would have to emerge from 
the Stone Age. Mindlin commanded and deserved respect—a former 
secretary of culture, he was one of Brazil’s leading intellectuals, 
owned the nation’s largest private library (with more than 20,000 
rare volumes), and ran a $200-million-a-year auto-parts business. The 
election of federation officers was approaching, Mindlin told me, and 
he had recommended that the organization’s president, Mario 
Amato, invite me to join his ticket. Amato agreed, but only, 1 later 
learned, after Mindlin had lobbied him and some other, less resistant 
directors. Maybe these other directors thought that, once enshrined 
in the Establishment, I would shut up. 

"I know what you think of the federation," Mindlin said. “But no 
one will listen to you while you're on the outside." 

I was dubious, until he mentioned that I would be the youngest 
director in the federations history and that we would work closely 
together to change things. I accepted immediately. 

My job was to oversee the technology department, which was 
responsible for such diverse activities as the development of industrial 
design in Brazil and the negotiation of international trade agree¬ 
ments. I soon learned that delegation was unheard of at the federa¬ 
tion. I had to personally approve purchases of office supplies, cash 
advances, absence forms—everything. There were always so many 
letters, memos, and forms awaiting my signature that 1 never got 
around to planning anything. When I would ask federation officers 

Maverick a 279 

why Joyce Leal, the department’s feisty, full-time executive, couldn’t 
sign them, their eyes would roll heavenward. 'And let ordinary 
employees make decisions?” they would say. 

One day I took some paperwork from my pile and gave it to Joyce 
“You sign these,” 1 told her, "and I’ll enclose a note saying this is the 
new arrangement in the technology department.” Joyce was a subver- 
sive at heart, too, so she agreed. 

We kept getting the documents back, unapproved, in the interof¬ 
fice mail. Mindlin and I even had to dig into our own pockets so the 
departments employees could travel, since our cash advance forms 
were bouncing along with everything else. But we didn’t give in. In 
fact, my war with the federation bureaucracy escalated. The depart¬ 
ment needed a new typewriter, but we knew it would take months of 
paperwork even if we played by the rules. So Mindlin sent us one 
from his company, while we waited for our request to clear channels. 
It took so long the borrowed typewriter broke. We had it repaired 

and sent the bill to the federation, which had the nerve to send it to 

Under the federation’s rules, employees who arrived late lost a 
chunk of pay. Was it any wonder that many were loath to stay even a 
minute late? We wanted our department’s employees to come to 
work at a time that suited them, so we changed their contracts to 
make them part-time workers, exempting them from the federation’s 
time cards. Of course, they still worked the same number of hours 
and were paid the same salary. But now they could start when they 

wanted. Just as with our workers at Semco, their motivation and 
their productivity grew. 

Even within our department, communication was in a shambles, 
in large part because almost everyone was tucked away in small 
offices and cubicles and had only a vague idea of what colleagues were 
doing. So we tore down all the walls and introduced everyone to one 
another. Then we transferred some employees who didn’t fit with our 
new style to other federation departments, and dismissed a few who 
weren’t up to par. But we didn’t replace anyone. That gave us more 
money to pay those who remained, who were so productive they 
more than made up for their former colleagues. 

These and other changes were provoking much gnashing of teeth 

280 A Ricardo Semler 

within the pyramid, but the federation couldn’t get rid of me. What 
I was doing was working. Companies all over Brazil were telling 
Amato how much our department had helped with this or that 
problem. Seminars we staged with international experts on technol¬ 
ogy management filled the federation’s meeting rooms. Our depart¬ 
mental reports and position papers were being quoted by government 
policymakers and legislators. 

Meanwhile, I followed my usual approach to leadership and backed 
away from the department as it revived. There would be no cult of 
personality here. After a few months I no longer signed any papers. I 
started coming in only once a week, then every other week, then 
once a month. Of course, no one noticed. Once I had gotten the 
bureaucracy off their backs, Joyce and the others didn’t need me. 

Two years later, it was time for new elections at the federation. 
Several directors tried to convince Amato to remove me from my spot 
at the bottom of the ticket, but he again stuck w'ith me. I ended up 
as a vice president, which confirmed a lesson I had learned through 
the years of change at Semco: it’s always better to seek forgiveness 
than ask for permission. 

A bit later, I received even more of an endorsement from the 
business community. Each year, The Gazeta Mercantile Brazil’s Wall 
Street Journal, and a group of leading magazines poll 54,000 compa¬ 
nies carefully chosen to represent the Brazilian economy, asking 
executives to vote, secretly, for Business Leader of the Year. It is a 
respected survey—or at least it was until 1990, when I finished first. 

From outcast to Establishment in a single bound. 

I give fifteen to twenty lectures a year—it would be more, but I 
charge an exorbitant fee to keep down the number—to audiences 
ranging from executives at such giants as General Motors, IBM, 
Lever Brothers, and Philips to partners of two-person franchising 
firms. I also speak at events as diverse as a Canadian telecommunica¬ 
tions convention and a gastroenterologists’ congress in Rio. 

Almost every lecture is followed by a stream of questions, about 
half from people who, like the doctor, are looking for lessons to 
appiy. That s also what brings so many executives to our door. 

We don t disappoint them. The programs we have developed at 

Maverick a 281 

Semco can be instituted in supermarkets, ad agencies, steel mills, 
baseball teams, and universities. Banks have implemented the reverse 
evaluation. In Brazil state-owned power utilities have begun to 
consult with employees before filling midlevel jobs. A bank in Rio 
has profit sharing, Semco-style. A General Motors subsidiary, Delco, 
practices reverse evaluation. A museum has democratic decision 
making. A hospital has open offices for administrators. And a 
classmate of mine at Harvard, Noel Ginsburg, reorganized his 
company, Container Industries, Inc., of Denver, Colorado, on the 
basis of discussions we had in our dorm. Noel gives his workers the 

same kind of information and freedom we do, and has been delighted 
by the results. 

Moreover, I believe military leaders would gain much from open 
discussions with the troops, although I am obliged to report that 
there were some frowns in the audience when I spoke to a group of 
many-starred generals at Brazil’s Superior School of War. Cutting 
bureaucracy would give religious leaders more time for their flocks, 
too. And, of course, industries in the burgeoning service sector could 
benefit from the increased efficiencies that follow when employees 
have a say in their work life. My stint at the federation proved that. 

But flattered as I am by all the companies that have imitated us, it 
makes me a bit nervous. For starters, I’m the first to acknowledge 
that Semco has a long way to go. We need to push decision making 
deeper into the organization, especially on matters of business 
strategy. Our employees need to understand our balance sheets better, 
too. We would like to have worker representatives on our corporate 
board, with full voting power, not as second-class citizens, as in some 

European countries. And we want everyone at Semco to set his own 

Even if we were perfect, no company should set out to copy us, 
program by program. Yes, headline memos cut paperwork and 
flextime improves productivity. Programs like job rotation, self-set 
salaries, and reverse evaluation will, I believe, improve any company, 
but that isn’t all there is to Semco. It would make me less like a 
snake oil salesman if, along with our programs, companies also 
adopted the philosophy of freedom and trust that inspires them. 
Modern managers are searching. Small companies are emulating 

282 a Ricardo Semler 

those large ones on the assumption that they must have done 
something right to get that big. Huge corporations, for their part, 
are envious of small and agile competitors. Entire American indus¬ 
tries look with both angst and hope to the Pacific Rim, even as Asian 
companies become more Westernized. 

In their desperation for quick fixes, too many executives are too 
quick to jump on the latest managerial fads and fashions, as if they 
will be panaceas for sagging productivity. Quality Circles. Just-in- 
time deliveries. Kanban production systems. Networking. Direct 
costing. Total quality, \bu ve heard all the buzz words. 

Transporting Asian values to, say, Smyrna, Tennessee, is like 
wearing a kimono to a Tupperware party. Nothing is less Western 
than the notion of total loyalty to a company, except possibly the 
belief that age should come before competence. If you must borrow 
from Japan, don’t forget to fill a 747 with enough Japanese to 
populate your factory. 

Very early one morning I visited the Ishikawajima-Harima ship¬ 
yard in Rio and found all the employees in the yard, singing the 
company song and exercising together. I pinched myself to make sure 
I wasn t in Kyoto. They had transplanted an activity that had taken 
2,500 years to evolve to a young plant on another continent. If by 
succeeding you mean that half the people at the shipyard were 
exercising because they thought it was the best route to a promotion, 
then it was a success. If the goal was to integrate the employees, 
build respect for their leaders, and mediate on the importance of the 

company in each worker s life, then they are all on the wrong side of 
the ocean. 



I know a textile company that wove fine English woolens. Its 200 
employees worked in a machine-filled factory set in what might be 
called an exurban industrial park. The chief executive was definitely 
performance-oriented, starting with his own: he arrived early, left 
late, and made all the important decisions in between. The factory 
was subdivided into specialized areas of production, each with its 
own boss. Each boss, in turn, had a group of foremen to watch the 
workers. Accountants and salespeople were on the mezzanine above 
the shop floor and reported to their respective department heads. 
Everything was strictly hierarchical and pyramidal. 

As I described this company to an international telecommunica¬ 
tions convention not long ago, I could see people in the audience 
becoming more and more puzzled. What was the point? they seemed 
to be wondering. It was just an ordinary business. There didn’t seem 
to be anything distinctive about it. 

Except that this textile factory existed in 1633. And the moral of 
the story: our advances in technology have far outstripped our 
advances in mentality. 

It was a particularly appropriate message for an audience of people 
who turn over our technological base every year or so. They have 
made it possible to instantly teleconference with China and call home 
from the belly of a 747 over the Pacific. And yet most businesses 
today are still organized much the same way they were in 1633, with 

284 A Ricardo Semler 

stultifying top-down management, close and distrustful supervision, 
and little room for creativity. The conflict between advanced technol¬ 
ogy and archaic mentality is, I believe, a major reason why the 
modern workplace is characterized by dissatisfaction, frustration, 
inflexibility, and stress. 

If only minds were as easy to change as machines. I’ll wager that 
it’s easier to invent a new generation of microchips than get a 
generation of middle managers to alter the routes they drive to work 
every day. Technology is transformed overnight; mentality takes 
generations to alter. Who can blame us for thinking technology will 
cure all that ails the workplace. It’s so much easier to acquire. 

Computer networks, automated machinery, rapid new product 
development, and quick and efficient communications all make a 
business up-to-date, but nothing more. And what is up-to-date? 
Rushing to the Caribbean for a one-week break, sharing sardine-can 
passenger compartments with 300 overstressed office workers, all 
eating from plastic containers, calling the company on the airphone, 
and dreaming of the languid turquoise waters and pristine beaches 
they will have to share with the occupants of ten other jumbo jets 
landing ahead of theirs. 

There s no doubt in my mind: technology has gone through the 
roof since 1633, but quality of life has gone down the drain. All we 
have done is accelerate our malfunctions and increase the intensity of 
our miscommunication. 

Let me propose a new definition: the truly modern company avoids 
an obsession with technology and puts quality of life first. 

It took us a while to see it clearly, but at Semco we now focus on 
innovations that will enable us to work better together, rather than 
simply trying to acquire the next generation of plasma welding 
computers or CAD/CAM machining centers. 

While I am pontificating, let me go a step further: no company 

can be successful, in the long run anyway, if profits are its principal 

It s gotten to the point that businessmen forget their own identi¬ 
ties and become what they invoice. I recently attended a luncheon in 
Sao Paulo with thirty industrialists who wanted to form a committee 

Maverick a 285 

to elect an alternate to the Federation of Industries board 
As all these powerful men were settling in around a U-shaped table, 
someone suggested that we introduce ourselves. 

The first rose and said: “My name is so-and-so, from company 
such-and-such, and we sell $200 million a year and employ 2,000 
employees." Then he sat down. Then the next person followed suit. 
And so it went. It was funny to watch these small men with large 
cigars getting up to announce proudly that they had double the sales 
of the person before them, or dejectedly acknowledge that their 
predecessor had beaten them by a few hundred million. You can 
imagine the discomfort when the smallest of the lot, barely a 
five-looter, declared that his company's sales had been $2.5 billion. 
It was all I could do to keep from giggling as I watched the awe and 
envy on my colleagues' faces. Moments later someone with a calcula¬ 
tor declared that together we represented sales of $5 billion. Never 
mind that our candidate for the federation board lost. 

How important is money, anyway? Entrepreneurs say profits are 
their raison d etre. In the months I spent at Harvard, I met many 
industrialists who, like me, were taking the program in annual 
installments. During the first session I noticed that some of them 
paid close attention to the personal finance course. They clearly had 
thoughts of selling their businesses. 

In the intervening year a few did cash in, trading their companies 
for a chance to realize dreams of an island, a sailboat, and maybe 
even golf every day. But by the time the second session at Harvard 
came along a few of these liberated souls were already a bit bored. 
And by the third session nearly all those who had sold their 
businesses had started or acquired new ones and were happily back on 
the job. They were never in business for the money. 

Neither was my father. He had long planned to retire at age 
fifty-five. As it approached, however, he moved the deadline to sixty, 
then sixty-five. By the time I started working at Semco he had 
slipped the date to his seventieth birthday. 

Nothing is more comprehensible (although I didn’t understand it 
at the time). Income, though vital, isn’t a satisfying end in itself. As 
my Harvard classmates discovered, you can sell your company for 
twelve or sixteen times earnings and not be happy. 

286 A Ricardo Semler 

At the risk of sounding patronizing, money isn’t the only goal of 
workers, either. We try to pay people at Semco more at Semco than 
they would receive elsewhere, and of course there’s profit sharing. 
But that isn’t why so few of our people leave. We offer employees a 
chance to be true partners in our business, to be autonomous and 
responsible. That s why, with some regularity, many of our key 
people regularly spurn lucrative employment offers. 

If money isn t all it’s made out to be, information is, I believe, a 
most undervalued commodity. There is power in knowing something 
someone else doesn’t, which explains why executives are so often 
loath to share information with employees. 

You don’t believe me? Try this test your next meeting. When some 
important item comes up, say that it s better not to discuss it now 
because a related issue was decided that same morning that will 
change things considerably. Then, after a few* moments of apprehen¬ 
sive silence, say that you just can’t discuss it now. 

I d be surprised if you haven’t just become the most powerful 

person in the room. You know something the others don’t—at least 
that’s what they think. 

But when cards are held close to the vest, communication will be 
faulty, and anxieties, misunderstandings, insecurity, and eventually 
hostility will manifest itself. No amount of “we’re-all-in-this- 

together-because-we re-all-one-big-family” sloganeering will compen¬ 

Which is why when we started sharing information at Semco it 
had such a profound effect. People in the higher echelons could no 
longer rely on the conventional symbols of power and had to develop 
leadership skills and knowledge to inspire respect. The centers of 
power shifted, as people that were formerly quiet and apolitical rose 
in stature and the people who were good at hobnobbing and 
gossiping eventually left. (And when they did, despite some pertur¬ 
bation from those who remained, we posted honest bulletin board 
notices explaining the reason for their departure.) 

One of the biggest misconceptions about modern man is that he is 
somehow different from his ancestors. Man has always lived in tribes 

Maverick a 287 

and I daresay always will. Whether these groups are ethnic, reli¬ 
gious, political, or vocational, they are our anchors. Being a 
Buddhist, a member of the National Rifle Association, a bird¬ 
watcher, a Nintendo fanatic, a Rotarian, or a knight in the Ku Klux 
Klan gives us an identity, for better or worse. 

People derive identity from their companies, too, wearing 
Mitsubishi or Motorola like a surname. And within companies they 
can belong to subtribes, each with their own norms of dress and 
conduct. Just as you will never confuse an Orthodox Jew with a Hare 
Krishna, never will finance executives, with their suspenders and 
Ferragamo ties, be mistaken for production guys, with their multiple 
pens safely tucked in plastic shirtpocket protectors. 

Companies and organizations must be redesigned to let tribes be. 
They must develop systems based on coexistence, not on some 
unattainable ideal of harmony. Different tribes will never fully 
integrate, which is why it is folly to try to create a “we’re-all-one- 
big-family atmosphere in the workplace. Fixed working hours, 
organization charts, and policy manuals are all so negative. They 
strip away freedom and give nothing in return but a false feeling of 
discipline and belonging. They elevate bureaucrats and ennoble 
conformity. By all means establish and promote a common goal, but 

recognize divergence and let people determine their own ways of 
achieving it. 

The issue of tribal coexistence is, I believe, critical for survival in 
modern times. Up until now it has been easy enough for the First 
World to keep its distance from the Third World and view the Southern 
Hemisphere as very far away. But technology is drawing everyone and 
everyplace close. Like lava from a huge volcano, tribes are moving 
toward areas where the standard of living is higher. In a few decades 
all that will be left of the First World will be a few ghettos of the 
super-rich, islands of luxury surrounded by misery. There will be a 
lot of Cairo in Paris, Mexico in Colorado, and Syria in Switzerland. 
And as the Third World makes its glacial movement north, it will 
leave behind places like Somalia, Bangladesh, and the Ivory Coast, 
which will become an even more abject Fourth World. 

Most companies and organizations are unprepared for this new 
world order. Their first impulse is to try to homogenize everything. 

288 A Ricardo Semler 

to set up elaborate training programs and teach immigrants to say 

have a nice day and watch your step in the hope they will 
eventually make baseball fans out of them. As if two Vietnamese 
clerks, a Hong Kong accountant, and six Canadian factory workers 
could work together, just like that. Consider the so-called guest 
workers in Germany and France. Even if Turks could speak flawless 
German, would they be accepted? I think most people want to be 
generous and egalitarian, but if they believe their family, their job, 
or their neighborhood is threatened, their tolerance for tribal coexist¬ 
ence evaporates. 

Discrimination will always exist, since it is tied up with tribalism, 
but there is plenty that can be done to diminish its effects. At 
Semco, we have built an organization without first- and second-class 
citizens. Our workers wear collars of many colors, not just blue and 
white, and we don t condone symbols of power or exclusivity such as 
executive cafeterias or reserved parking places. 

I know the coexistence of tribes is possible. I have witnessed it to a 
degree I would not have suspected. In 1992, a demonstration for the 
impeachment ol Brazilian President Collor brought tens of thousands 
of workers to a large square in the center of Sao Paulo. I was 
watching from the sidelines when I was spotted by a director of the 
radical Workers Party, who called me to the podium and insisted I 
speak to the demonstrators, some of whom were waving signs 
reading down with business and worse. After I finished, I got an 
embrace from the party’s president. 

There I was, Brazil’s Business Leader of the Year, being cheered by 
the most extreme of workers. 

In the 1920s, an engineer named De Forest went to see Harry 
Warner, of Hollywood’s famous Warner Brothers. De Forest had 
managed to synchronize sound and image and could, he said, 
transform silent movies into talking pictures. Warner listened to 
him, then replied: ‘Are you crazy? Who wants to hear an actor talk?” 

Henry Ford would sell his Model Ts in any color, as long as it was 
black. Legend has it that he adopted this monochromatic philosophy 
to simplify production and keep prices down. To this day Ford is 
regarded as a marketing hero. I have a revisionist view. Old Henry’s 

Maverick a 289 

stubborn thinking cost Ford the leadership of America's biggest 
industry, for William Durant, of the then smaller General Motors, 
decided to offer cars in a variety of colors, and was soon looking at 
Ford through his rearview mirror. 

And then there was Chester Carlson, who visited IBM, GE, and 
RCA, trying to sell his new invention. They thanked him for his 
time but regretted that, in their view, his idea had no future. A 
stubborn man, Carlson persisted. Finally, he met Joseph Wilson, who 
owned a small firm named the Haloid Company. Wilson saw the 
potential of Carlson’s device. With time, he even renamed his 
company after it: Xerox. 

Corporations have notoriously short life spans. Even in the stable 
and relatively prosperous United States, a company has a less than 5 
percent chance of being in a better position fifty years from now. 
These cautionary tales illustrate what I believe is the biggest chal¬ 
lenge any business faces: change. Semco has succeeded despite some 
of the harshest economic conditions imaginable because we have 
learned to see the need for change and have been smart enough to 
seek our employees help in implementing it. 

To survive in modern times, a company must have an organizational 
structure that accepts change as its basic premise, lets tribal customs 
thrive, and fosters a power that is derived from respect, not rules. In 
other words, the successful companies will be the ones that put 
quality of life first. Do this and the rest—quality of product, 
productivity of workers, profits for all—will follow. 

At Semco we did away with strictures that dictate the “hows” and 
created fertile soil for differences. We gave people an opportunity to 
test, question, and disagree. We let them determine their own 
training and their own futures. We let them come and go as they 
wanted, work at home if they wished, set their own salaries, choose 
their own bosses. We let them change their minds and ours, prove us 
wrong when we are wrong, make us humbler. Such a system relishes 
change, which is the only antidote to the corporate brainwashing 

that has consigned giant businesses with brilliant pasts to uncertain 

People often assume that at Semco we have strict selection proce¬ 
dures to insure that those who come to work for us are philosophi- 

290 a Ricardo Semler 

cally attuned with our system. In fact, it’s the other way around. We 
look lor competence, and ignore all else. Many of our employees 
regularly question our concepts. There are even pockets at Semco 
that are autocratic, and people who like to work in that kind of 
environment have slowly migrated there. But how can we lock out 
people who don t think the same way we do without becoming 

people who say things like, “This is not the way we do things around 

Semco is more than novel programs or procedures. What is 
important is our open-mindedness, our trust in our employees and 
distrust of dogma. We are neither socialist nor purely capitalist, but 
we take the best of these failed systems and others to reorganize work 
so that collective thinking does not overpower individualistic flights 
of grandeur; that leadership does not get lost in an endless search for 
consensus; that people are free to work as they like, when they like; 
that bosses don t have to be parents and workers don’t act like 
children. At the heart of our bold experiment is a truth so simple it 

would be silly if it wasn’t so rarely recognized. A company should trust 
its destiny to its employees. 

No, Semco isn’t a model, with programs to be followed with 
precision, so many recipes for participation, productivity, and profits. 
Semco is an invitation. I hope our story will cause other companies to 
reconsider themselves, and their employees. To forget socialism, 
capitalism, just-in-time deliveries, salary surveys, and the rest of it, 
and to concentrate on building organizations that accomplish that 
most difficult of all challenges: to make people look forward to 
coming to work in the morning. 






I promised in Chapter 9 to reveal my cure for time sickness. (All 
eyes on the top hat, please. Here comes the rabbit.) 

1. Begin at the end. Set a certain hour at which to leave the office 
and obey it blindly. I chose 7 p.m., but before I had often worked 
until midnight. If you normally work until 7 p.m., move up 
quitting time to 5:30 or 6:00. If you take work home on weekends, 
establish a sixty-day program to halt this insidious practice. 

2. Sift though that stack of papers on your desk and decide which 
are most important. (Deciding that everything is equally important 
is cheating. Start again.) Spend several hours, or even a whole day, if 
that’s what it takes, discovering what’s in that pile. Begin with the 
most difficult, complex, or time-consuming documents. In other 
words, go through the pile in order of importance, not appearance. 
You won t get a false sense of accomplishment that way. As you go 
through it, divide the papers into three categories: 

Priority items, which require your personal attention and 
represent matters of indisputable importance. Don’t put more 
than five documents in this category. 

Items that can be handled only by you, can wait. At first this 
category seems the most enjoyable, because so much appears to 

294 A Ricardo Semler 

fit in it. Think hard about whether you are really the only one 
who can deal with an item. Whether your subordinates or 
colleagues are overworked should not weigh in your decision; the 
control of time is nothing if not an exercise in selfishness. Load 
them up with everything that fails “The Test of the Seventies.” 
(No, this isn t a quiz. See if you remember Watergate.) Ask 
yourself: Is it possible that someone else could do this task at 
least 70 percent as well as I could?” If the answer is yes, let him. 
Items you think would be good to look at, but never quite 
get around to. These include newspapers and magazines, 
lengthy reports, copies of memos—you get the idea. We’ve all 
grown accustomed to receiving vast quantities of information. 
As a defense, we tend to read a little of everything. This is 
among the most serious causes of time illness. 

The key to time management is self-esteem. You must maintain it 
even though you may not be as well versed about some essentially 
meaningless report or arcane issue as your associates. You must be 
prepared to go to a meeting and endure comments such as, “You 
mean you didn t read Short-cuts to a Better Casting Project’ in the 
latest Foundry Lovers' News? Better to suffer the humiliation of saying 
you didn t and ask someone to please be kind enough to summarize it 
than to have had to read all the articles that cross your desk. Legions 
of executives believe they will be regarded as ill-informed dunces if 
they let their subscriptions lapse to The Wall Street Journal, The 
Financial Times, Newsweek, Time, and LExpress, not to mention 
assorted local papers, financial newsletters, Fortune, Forbes, Business 
Week, and so on. Publications work hard to convey their indispens¬ 
ability. Don’t let too many succeed. 

Due to (what else?) lack of time, I would leave magazines and 
newspapers on a table in my office to read later. When more 
periodicals would arrive, I would carefully add them to the pile. I 
felt depressed just looking at it, fearful of all the information I had 
not absorbed. Then one evening the pile collapsed. That was my cue, 

and 1 finally gave the magazines the dignified burial they deserved, 
without opening them. 

1 estimate that the ratio of useless to relevant reading material is 

Maverick a 295 

about twenty to one. With that in mind, my advice is to reduce the 
literary inflow to a maximum of two newspapers a day, two weekly 
magazines, and two publications in a specialized field. Start being 
proud of not being aware of everything. Get off distribution lists. 
The reward will be an opportunity to engage in that underappreciated 
occupation, contemplation. Aristotle, who didn’t subscribe to The 
Wall Street Journal , once said, "Thinking requires leisure time.’’ If 
you are not in possession of leisure time, you can’t be thinking all 
that much. 

3. If you thought you could manage your time without making 
any investment in fixed assets, you are wrong. There is one essential 
acquisition: another trash can. 

I know. You already have a perfectly good one. But most people 
have enough on their desk to fill two. At first, using this second 
trash can will take courage. Y)u will have to toss into it impressive 
reports and unread magazines. But remember the question Alfred 
Sloan, the legendary head of General Motors, used to ask: "What is 
the worst thing that can happen if I throw this out?’’ If you don’t 
tremble, sweat, or grow short of breath, go ahead and pitch it. 
Eventually, your second trash can will become a baby-sitter for your 
in-box. Leave it in place a few more months, a magnificent symbol of 

4. Think hard before accepting that invitation to lunch or to visit 
a supplier or to make a speech to a trade group. The first response is 
usually to reach for the Filofax and, if the appropriate space is blank, 
scrawl in the new commitment. Adopt a new party line: "Thanks, 
but I just can’t fit it in.’’ Or, you might ask, "Have you tried X?’’ 
Or: "I can’t make it, but do let me know what happened.’’ Or, as a 
last resort: "I’m really sorry, but I’ll be on my honeymoon.’’ Say what 
you have to, but take part only in events that are absolutely, 
positively essential. That lunch to get to know supplier Z better or 
to impress customer Y, like those "it’s-probably-a-good-idea-to-be- 
there’’ meetings, play upon insecurities. I’ve never met a supplier 
who lowered his price because he found me great company at lunch 
or a customer who gave me an order because the Beaujolais I served 

296 A Ricardo Semler 

at dinner made him giddy. Participate only in what you’re sure will 
generate a return for your investment of precious time. 

11 you have to meet with someone, do it at the office. The 
surroundings discourage people from drifting from the subject. But 
don’t provide visitors with a cup of coffee: it’s an invitation to an 
easygoing, unproductive conversation. By leaving an intruder 
uncaffeinated, you are contributing to his health, and he may not be 
so quick to return, either. 

5. Rationalizing time without talking about meetings is like a 
soccer match without a riot. Is there anyone who works in an office 
who doesn t go to too many meetings? 

For starters, remember that man is a social animal. We feel more 

comfortable in the presence of others. Meetings give rise to the sense 

of being part of a group, of solidarity even with the ever-present rifts 

and jealousies of business. And, of course, corporate communication 

systems are usually slow and often inadequate, so meetings let us 
catch up. 

To eliminate meetings is to go against human nature and diminish 
corporate efficiency. But making meetings more effective is not all 
that difficult. Through trial and error, we at Semco have come up 
with some recommendations. 

Begin on time. (Five or ten minutes late is still on time, 

unless you’re Swedish.) Just start with whoever is there. Do 

this a few times and those who usually are late will get the 

Dont start a meeting without first setting a time to stop. 
And dont go beyond it by more than a few minutes. When 
you are in your own office, get up from your chair and say, 
“Okay then,” when you want to end a session. Sometimes I 

sit on the edge of my desk from the start. It may not be 
polite, but it works. 

• Go over the agenda in front of everyone. List subjects in order 
of importance. Don’t give in to the temptation of clearing up 
old items, or getting rid of uncomplicated new items, first. 

• Delegate to one or more people any item that might take 

Maverick a 297 

more time than is allotted for it, or that provokes a discussion 
that drags on without hope of a resolution. 

• Don’t have meetings that last longer than two hours. After 
that, attention drops by the minute. 

Avoid dog-and-pony shows. Keep reports short and discourage 
the use of charts and tables. Avoid overhead transparencies, 
too, and don’t ever turn out the lights. 

• Be a bear about interruptions. The only excuse for breaking 
into a meeting is a customer with a problem. 

• Transform as many meetings as possible into telephone calls or 
quick conversations in the hall. People tend to call meetings 
for problems that can be resolved in ten or fifteen minutes 
over the phone, or even in a fax. 

6. About the telephone: Anyone who takes a message at Semco 
asks the caller to detail the subject—that’s detail, not state. Ask your 
secretary or assistant to automatically say that you cannot take a 
call—before the caller is asked who he is, of course. Take the list of 
callers from a given day (or several days) and return only those from 
people you truly need to talk to, such as customers. As for the 
others, they will either: 

• Call again and again, until they give up, at which point you 
know the subject was not important. 

• Call one of your associates. 

In any case, prepare to hear how talking to you is harder than 
talking to the Pope. Take it as a compliment. 

7. Make time to think. Try blocking out a half day a week on 
your agenda. I find that Monday and Friday mornings are good, 
because I can clear away post- and pre-weekend distractions. During 
this half day, avoid your office. Camp out in an unused conference 
room or, even better, stay home. 

Thinking is difficult. It requires concentration and discipline. Give 
it the time it deserves. Aristotle would approve. 




What follows are excerpts from the small booklet we give to each new Semco 
employee. It constitutes our only written set of rules. 

300 ▲ Ricardo Semler 


nf . Se . m . C ° d ° eSn C USe a formal organization chart. Only the respect 
of the led creates a leader. When it is absolutely necessary to sketch 
the structure o, some part of the company, we always do it in pencil, 
and dispense with it as soon as possible. 

Cartoons by Miguel Paiva 

Maverick a 301 


When people are hired or promoted, the others in that unit 
have the opportunity to interview and evaluate the candidates before 
any decision is made. 

302 A Ricardo Semler 


Semco has flexible working hours, and the responsibility for 
setting and keeping track of them rests with each employee. People 
work at different speeds and differ in their performance depending on 

the time of day. Semco does its best to adapt to each person s desires 
and needs. 

Maverick a 303 


We want all our people to feel free to change and adapt their 
working area as they please. Painting walls or machines, adding 
plants or decorating the space around you is up to you. The company 
has no rules about this, and doesn’t want to have any. Change the 
area around you, according to your tastes and desires and those of the 
people who work with you. 

304 a Ricardo Semler 


Unions are an important form of worker protection. At Semco, 
workers are free to unionize and the persecution of those connected 
with unions is absolutely forbidden. Unions and the company don’t 

always agree or even get along, but we insist that there is always 
mutual respect and dialogue. 

Maverick a 305 


Strikes are considered normal. They are part and parcel of 
democracy. No one is persecuted for participating in strikes, as long 
as they represent what the people of the company think and feel. The 
workers’ assemblies are sovereign in this respect. 

Absence at work for reasons of strike is considered as normal 
absenteeism, without further consequences or punishments. 

306 a Ricardo Semler 


Our philosophy is built on participation and involvement. 
Don’t settle down. Give opinions, seek opportunities and advance¬ 
ment, always say what you think. Don’t just be one more person in 
the company. 

Your opinion is always interesting, even if no one asked you for 
it- Get in touch with the factory committees, and participate in 
elections. Make your voice count. 

Maverick a 307 


Twice a year you will receive a questionnaire to fill out that 
enables you to say what you think of your boss. Be very frank and 
honest, not just on the form but also in the discussion that follows. 

308 A Ricardo Semler 


Employees at Semco are guaranteed representation through the 
Factory Committee of each business unit. Read the charter, partici¬ 
pate, make sure your committee effectively defends your interests— 
which many times will not coincide with Semco’s interests. We see 
this conflict as healthy and necessary. 

Maverick a 309 


Many positions at Semco carry with them hierarchical authority. 
But efforts to pressure subordinates or cause them to work out of fear 
or insecurity, or that show any type of disrespect, are considered an 
unacceptable use of authority and will not be tolerated. 

310 A Ricardo Semler 



Anyone who has been with us for three years, or has reached the 
age of fifty, has special protection and can only be dismissed after a 
long series of approvals. This doesn’t mean Semco has a no-layoff 
policy, but it helps to increase the security of our people. 

Maverick a 311 


Semco is a place where there are big changes from time to time. 
Don’t worry about them. We consider them healthy and positive. 

Watch the changes without fear. They are characteristic of our 

312 a Ricardo Semler 



Neither has any importance at Semco. A person’s appearance is 
not a factor in hiring or promotion. Everyone knows what he or she 
likes or needs to wear. Feel at ease—wear only your common sense. 

Maverick a 313 


Everyone owns his or her life, and a persons private affairs are 
considered sacred by the company. Semco has no business interfering 
with what people do when they are away from work, as long as it 
doesn’t interfere with work. Of course, our human resources depart¬ 
ment is at your service for any help or support you may need. 

314 A Ricardo Semler 


The company loans money to its people in unforeseen situa¬ 
tions. Thus, loans to cover purchases of a home, a car, or other 
predictable expenses are not included. But the company wants you to 
know that the day you run into a difficult and unexpected financial 
problem it will be there for you. 

Maverick a 315 


It’s only worth working at a place in which you can be proud. 
Create this pride by insuring the quality of everything you do. Don’t 
let a product leave the company if it’s not up to the highest 
standards. Don’t write a letter or a memo that is not absolutely 
honest. Don’t let the level of dignity drop. 

316 A Ricardo Semler 



Semco and its people must strive to communicate with frank¬ 
ness and honesty. You must be able to believe fully in what is said to 

you by your co-workers. Demand this transparency when you are in 

Maverick a 317 


Promoting a birthday party at the end of the workday, barging 
into meetings where you were not invited, or using nicknames are all 
part of our culture. Don’t be shy or stick to formalities. 

318 A Ricardo Semler 


Semco doesn't believe in giving prizes for suggestions. We want 
everyone to speak out, and all opinions are welcome, but we don't 
think it is healthy to reward them with prizes or money. 

Maverick a 319 


Women in Brazil have fewer employment, promotion, and 
financial opportunities than men. At Semco, women have various 
programs, run by women, that seek to reduce this discrimination. 
They are known as The Semco Women. 

If you are a woman, participate. 

If you aren’t, don’t feel threatened and don’t fight against this 
effort. Try to understand and respect it. 

320 a Ricardo Semler 


Semco is not one of those companies that believes anyone is 
irrep aceable Everyone should take their thirty days of vacation every 
year. Its v.tal for your health and the company’s welfare. No excuse 
good enough for accumulating vacation days for "later.” 



The following questionnaire is anonymously completed by all 
Semco employees every six months as part of the process of evaluat¬ 
ing their supervisors. The questions are weighted according to their 
importance and the results are posted. A score of 80 out of 100 is 

1. When an employee makes a small mistake, the subject is: 

a. Irritated and unwilling to discuss the mistake 

b. Irritated but willing to discuss it 

c. Realizes the mistake and discusses it in a constructive manner 

d. Ignores the mistake and only pays attention to more important 

2. The subject reacts to criticism: 

a. Poorly, ignoring it 

b. Poorly , rejecting it 

c. Reasonably well 

d. Well, accepting it 

322 A Ricardo Semler 

3. The subject is: 

a. Constantly tense 

b. Usually tense, but relaxed on occasion 

c. Usually relaxed , but tense on occasion 

d. Constantly relaxed 

4. The subject is: 

a. Insecure 

b. More often insecure than secure 

c. More often secure than insecure 

d. Secure 

5. As far as professional and personal relationships are concerned, 
the subject is: 

a. Incapable of separating them 

b. Frequently incapable of separating them 

c. Usually capable of separating them 

d. Capable of separating them 

6. When the subject’s department achieves a high level of produc¬ 
tivity, he or she usually: 

a. Takes credit for others' success 

b. Gives credit to those who did the work 

c. Gives credit to the team as a whole 

7. The subject is seen as: 

a. Always unfair 

b. More often unfair than fair 

c. More often fair than unfair 

d. Always fair 

8. The subject conveys to his/her team feelings of: 

a. Fear and insecurity 

b. Indifference 

c. Security and tranquility 

Maverick a 323 

9. The subject transmits to his/her team a sense of: 

a. Coldness and unwillingness to talk 

b. Distance, but willingness to talk 

c. Friendliness, but indifference to others’ problems 

d. Friendliness and concern with others’ problems 

10. When dealing with people in inferior positions (custodians, 
messengers, drivers, etc.), the subject usually: 

a. Has an attitude of superiority 

b. Ignores them 

c. Treats them politely, but with an air of superiority 

d. Respects them 

11. The subject treats his/her subordinates: 

a. Much worse than he!she treats superiors 

b. A little worse than he!she treats superiors 

c. Treats both the same 

12. The subject: 

a. Constantly reminds everyone helshe is the boss 

b. Occasionally reminds everyone hel she is the boss 

c. Rarely makes a point of being the boss 

13. The subject is: 

a. A weak leader, unable to motivate his/her team 

b. A weak leader, but able to motivate his!her team 

c. A strong leader, but unable to motivate his/her team 

d. A strong leader, and able to motivate his!her team 

14. When his/her team has a specific goal, the subject: 

a. Demands results, but doesn't participate in the effort to achieve 

b. Demands results, and participates superficially 

c. Participates in the effort when necessary to meet the goal 

324 A Ricardo Semler 

15. The subject: 

a. Is openly held in disrespect by his/her team 

b. Is held in disrespect by his/her team, but not publicly 

c. Generates neither respect nor disrespect 

d. Is respected by his!her team 

16. The subject: 

a. Gives obvious preferential treatment to some people because of their 
color, religion, or origin 

b. Denies being biased, but doesn't give equal opportunity to every¬ 

c. Isn’t biased and gives equal opportunity to everyone 

17. The subject: 

a. Gives obvious preference to people of a certain gender 

b. Dentes being biased, but doesn't give equal opportunity to every- 

c. Isn’t biased and gives equal opportunity to everyone 

18. When promotions and prizes are concerned, the subject: 

a. Gives them to those he or she likes 

b. Sometimes gives them to those who deserve them and sometimes gives 
them to 1 followers” 

c. Almost always is just and impartial 

19. During a crisis, the subject: 

a. Disrupts the group's unity 

b. Doesn t affect the group 's unity 

c. Helps the group stick together 

20. Which is more important to the subject: 

a. Work to be performed perfectly 

b. Work to be performed quickly 

c. Either speed or perfection, depending on the situation 

Maverick a 325 

21. The subject is: 

a. Excessively involved in all situations 

b. Not involved enough in all situations 

c. Adequately involved in all situations 

22. The subject’s knowledge of his/her area is: 

a. Insufficient 

b. Sufficient 

c. Profound 

23. If the subject were to replace you temporarily, his/her perform¬ 
ance would be: 

a . U nsatisfactory 

b. Regular 

c. Good 

d. Better than yours 

24. In choosing the urgent and the important, the subject: 

a. Doesn't know the difference between them 

b. Usually tends toward the urgent 

c. Distinguishes well between the two 

25. The subject: 

a. Wastes too much time on urgent problems 

b. Gives equal time to urgent and important matters 

c. Gives more time to important matters 

26. The subject is: 

a. Not very creative and resists neu> ideas 

b. Too creative and change-oriented , disturbing the atmosphere 

c. Is adequately creative and change-oriented 

326 A Ricardo Semler 

27. As far as creating an environment where people feel free to 
creative or suggest changes, the subject: 

a. Blocks innovative and creative ideas 

b. Doesn t block them, but also doesn't encourage them 

c. Promotes creative or innovative ideas 

28. As far as the team is concerned, the subject: 

a. Usually chooses the wrong people 

b. Sometimes chooses well and sometimes chooses poorly 

c. Usually chooses the right people 

29. The people who work around the subject: 

a. Rarely feel motivated to work 

b. Sometimes feel motivated to work 

c. Usually feel motivated to work 

30. The subject s use of financial resources given to him/her is: 

a. Poor 

b. Average 

c . Good 

d. Excellent 

31. The subjects use of his/her own time is: 

a. Bad 

b. Average 

c. Good 

d. Excellent 

32. The value the subject gives to framing and related matters is 

a. Too small 

b. Sufficient 

c. Great 

Maverick a 327 

33. The subject performs tasks: 

a. Almost always poorly 

b. Sometimes poorly and sometimes well 

c. Almost always well 

34. Regarding opinions that differ from his/hers, the subject: 

a. Never accepts them 

b. Usually doesn't accept them 

c. Sometimes accepts them 

d. Almost always accepts them 

35. People find the subject: 

a. U ntrustworthy 

b. Occasionally trustworthy 

c. Very trustworthy 

36. The subject represents the company: 

a. Poorly, raising concern about it 

b. Neither poorly nor well 

c. Well, leading people to trust it 




BOSSES: Semco doesn’t have as many as it used to, and those that 
remain aren’t very bossy. As workers began to exercise more control 
over their jobs, the need for supervisors diminished. We also have 
reduced our corporate staff, which provides legal, accounting, and 
marketing expertise to our manufacturing units, by more than 75 
percent, eliminating the data processing, training, and quality 
control departments, among others. 

CIRCULAR ORGANIZATION: We slashed Semco’s bureaucracy 
from twelve layers of management to three and devised a new 
organization structure based on fluid concentric circles instead of a 
rigidly hierarchical pyramid. All employees have one of only four titles: 
Counselors, who are like vice presidents and higher in conventional 
companies and coordinate our general policies and strategies; Partners, 
who run our business units; Coordinators, who comprise those first, 
crucial levels of management, such as marketing, sales, and production 
supervisors or engineering and assembly area foremen; and Associates, 
which is what we call everyone else. (See Chapter 24.) 

CLEAN-OUTS: Twice a year Semco shuts down for an afternoon 
and everyone cleans out his workplace. Office workers pitch useless 
files, factory workers get rid of scrap and old machinery. This 
program started in our offices, but now no one is exempt and nothing 
safe. (See Chapter 18.) 

330 A Ricardo Semler 

CORRUPTION: More than a few government inspectors who set 
out to extort Semco have ended up in jail. This has caused us plenty 
of trouble with the government inspectors all over Brazil, but it’s 
worth it rather than send a signal to our employees and customers 
that we tolerate dishonesty. (See Chapter 28.) 

DEMOCRACY: A cornerstone of the Semco system. We have 
representative democracy through our factory committees; on impor¬ 
tant decisions, such as plant relocations, all our employees get a 
direct vote. Subordinates essentially vote for their bosses. And to 
make the point absolutely clear, we don’t have executive dining 
rooms or reserved parking spaces. 

FACTORY COMMITTEES: At each Semco unit groups of 
workers machinists, office personnel, maintenance workers, stock- 
room personnel, draftsmen, and other types of employees except 
management—elect representatives to serve on committees. (The 
union is also represented.) These committees meet regularly with the 
top managers at each unit to discuss any and all workplace issues or 
policies. They are empowered to call strikes, audit the books, and 
question all aspects of management. (See Chapter 10.) 

FAMILY SILVERWARE: When a J0 b opening occurs or a new 

position is created, a Semco employee who meets 70 percent of the 

requirements is given preferential consideration over an outsider (See 
Chapter 22.) 

FLEXTIME: Many companies let office workers set their own hours. 
We are the only company I know of to extend this freedom to factory 
workers. This is potentially disruptive, since each shop-floor worker so 
heavily depends on his teammates, but our people have always subordi¬ 
nated individual preferences to group schedules. (See Chapter 30.) 

HEADLINE MEMO: All memos at Semco are limited to a single 
page, topped with a newspaperlike headline that gets right to the 

Maverick a 331 

point. There are no exceptions, not even marketing reports. (See 
Chapter 18.) 

HEPATITIS LEAVE: Our version of sabbaticals. Professionals can 
take a few weeks or even a few months every year or two away from 
their usual duties to learn new skills, redesign their job, or simply 
recharge. This program began when several harried managers told us 
they didn’t have time to think. We responded by asking them to 
consider what would happen if they suddenly contracted hepatitis 
and were forced to spend two months recuperating. Then we let 
them. (See Chapter 20.) 

JOB ROTATION: We encourage our managers to exchange jobs 
with one another, and in a given year up to 25 percent do. A 
minimum of two years and a maximum of five years in the same job 
are ample, but as with other programs at Semco it’s up to the 
employees to take the initiative. Job rotation can be unsettling, but 
offers some considerable advantages, including: 1) obliging people to 
learn new skills; 2) discouraging empire building; 3) providing 
employees with a broader view of a company, which lets them 
appreciate colleagues’ problems; 4) forcing management to prepare 
more than one person for a job; 5) creating additional opportunities 
for those who might otherwise be trapped. (See Chapter 20.) 

JOB SECURITY: There was a time not too long ago when IBM 
and Bridgestone Tire, 3M, and Kyocera, among others, boasted 
credos that read, “lifetime employment guaranteed here.’’ The Amer¬ 
ican executives of these companies would nod to their Japanese 
counterparts, and vice versa, through interviews in Fortune, The 
Financial Times, and Asahi Shimbun in which they would extol the 
high-minded strategy of “believing in people.’’ Now all this is legend 
and Japanese and American multinationals alike have laid off thou¬ 
sands of experienced workers. At Semco, we never promised anyone 
job security, so we never had to break this promise. 

LOST IN SPACE: Each year business is good we choose several 

332 A Ricardo Semler 

young people from our entry-level applicants and turn them loose. 
With no job description, no boss, no set responsibilities, they are 
free to roam, so long as they work in at least twelve departments in 
their first twelve months. Then they can negotiate a more permanent 
arrangement with any of those departments. (See Chapter 29.) 

signed to allow people to mingle. Our offices don’t have walls; people 
are instead separated only by plants. Sometimes we intentionally mix 

departments. In fact, we don’t like barriers of any kind separating 
our employees. 

MANUFACTURING CELLS: Instead of assembly lines, we clus¬ 
ter production machines so teams of workers can assemble a complete 
scale, dishwasher, mixer, or other product, not just an isolated 
component. That gives workers more autonomy and responsibility, 
which makes them happier and our products better. Nearly all 
factory workers have mastered several production jobs and some even 
drive forklifts to keep teammates supplied with raw materials and 
spare parts, which workers have been known to purchase themselves 
from suppliers. Often factory workers set production quotas and 
develop improvements for our products. (See Chapter 16.) 

NATURAL BUSINESS: A guiding principle. At Semco we have 
stripped away the unnecessary perks and privileges, such as executive 
dining rooms and fancy office furniture, that feed the ego but hurt 
the balance sheet and distract everyone from the crucial corporate 
tasks of making, selling, billing, and collecting. 

small group of employees, mostly engineers, who are free of any 
day-to-day production or managerial responsibilities and have all 
their time to invent new products, refine old ones, devise market 
strategies, unearth cost reductions and efficiencies, even dream up 
new lines of businesses. Their compensation depends in part on the 
success of their entrepreneurial efforts. (See Chapter 29.) 

Maverick a 333 

PATERNALISM: A dirty word at Semco. We don’t want to be a 
big, happy family. We want to be a successful business. We’re only 
concerned with our employees’ performance on the job, not their 
personal lives. You won’t find a running track, swimming pool, or 
gym at Semco. If our people want to join a health club, that’s their 
business. We do offer health insurance and other benefits, but we ask 
employees to help manage them. Occasionally, Semco will lend 
employees money, but only for unpredictable emergencies. Instead of 
treating employees like children who need looking after, we treat 

them like adults who are capable of making decisions on the job. (See 
Chapter 21.) 

PROFIT SHARING: Unlike typical plans, in which manage¬ 
ment unilaterally decides how much will be distributed and to 
whom, Semco negotiated with our workers over the basic percentage 
to be distributed about a quarter of corporate profits—and then 
they decide how to split it. (See Chapter 17.) 

REVERSE EVALUATION: Before anyone is hired or promoted to 
a leadership position, he/she must be interviewed, evaluated, and ap¬ 
proved by all the people who will work for him/her. Also, every six 
months Semco managers are evaluated by the people they supervise, 
who anonymously complete a multiple-choice questionnaire we 
developed lor this purpose. The grades are posted for all to see. There 
are no hard and fast rules, but those who consistently get poor marks 

(80 out of 100 is average) usually leave Semco sooner or later. (See 
Chapter 22.) 

RISK SALARY: About a third of all Semco employees have the 
option of taking a pay cut of up to 25 percent and then receiving a 
supplement raising their compensation to 150 percent of normal if 
the company has a good year. If Semco does poorly, on the other 
hand, they’re stuck with 75 percent of their salary. This program 
rewards those willing to take a risk, and lets some of our labor costs 
fluctuate with profits or losses. (See Chapter 25.) 

334 a Ricardo Semler 

RULES: We have as few as possible. There are no dress codes, for 
example, or regulations on travel. And we no longer have an internal 
auditing department to check up on our people. All we say is: use 
your common sense. (See Chapter 12.) 

SALARY SURVEYS: Many companies rely on off-the-shelf wage 
surveys to reevaluate salary scales. Semco is the only company I know 
of that has asked its factory and office workers to help in this process. 
They visit other manufacturing companies that are comparable to 
Semco, interview their counterparts, and tally the findings. As a 
result, no one distrusts our surveys. Eventually, though, we hope all 
Semco employees will set their own salaries. 

SATELLITE PROGRAM: An effort to contract out basic manu¬ 
facturing, with a twist: instead of doing business with outsiders, we 
help our own workers set up their own companies, transforming 
them from employees to partners. To ease the transition, we lease 
them our production machines at favorable rates and offer them 
advice on such matters as pricing, quality, and taxes. They are free to 
sell to our competitors, too. (See Chapter 32.) 

SELF-SET PAY: Paying people whatever they want seems a sure 
route to bankruptcy, but we’ve been doing this for years and 
we’ve never done better. A 10 percent raise turns out to be an 
exception. Nearly 25 percent of our employees now set their own 
salaries, including most of our Coordinators, and I don’t see any 

reason why factory workers can’t one day determine their own pay. 
(See Chapter 23.) 

SIZE: Large, centralized organizations foster alienation like stag¬ 
nant ponds breed algae. We believe people reach their potential only 
when they know almost everyone around them, which is generally 
when there are no more than 150 coworkers. When our business 

units grow bigger than that, we split them. (See chapters 14 
and 15.) 

Maverick a 335 

STRIKES: Regrettably, we haven’t eliminated them. But when 
there’s a walkout we: 1) treat everyone like adults; 2) tell the strikers 
that no one will be punished when they return; 3) don’t keep records 
of who came to work and who led the walkout; 4) never call the 
police or try to break up a picket line; 5) maintain all benefits; 6) 
don’t keep workers or union leaders out of the factory; 7) insist that 
everyone respect the right of those who want to work; 8) don’t fire 
anyone during or after the strike. (See Chapter 13.) 

SUPPORT STAFF: W^e have completely eliminated secretaries, 
receptionists, personal assistants, and other ungratifying, dead-end 
jobs from our payroll. In keeping with our philosophy of natural 
business, everyone at Semco fetches their own guests, copies their 
own papers, and sends their own faxes. (See Chapter 18.) 

TRAINING: Rather than have formal training programs, we 
respond in various ways when employees ask for a chance to develop 
new skills. We ask people to think about what they would like to be 
doing in five years and then prod them to request training that takes 
them there. The expenses are then approved at the weekly meetings 
of our business units. 

TRANSPARENCY: We make public virtually all corporate infor¬ 
mation, from salaries to strategies and productivity statistics to profit 
margins. We also give courses to teach workers to read financial 
documents such as balance sheets and income statements. Employees 
are free to question managers on any aspect of our business, and to 
talk to the media without fear. (See Chapter 21.) 

WORKING AT HOME: I do it at least three mornings a week, 
and encourage everyone else who can work at home to do so. It 
enhances concentration and productivity and gives people more 



the company’s financial statements, make cor- 
porate decisions, take over the cafeteria 
kitchen, start their own business with company 
assets, and redesign the products Semco builds 
and how they are built. The result: in ten years 
of constant experimentation—a time period 
in which Brazil’s economy faltered—Semco 
has achieved a growth rate of 600 percent. 

In chronicling Semco’s inner revolution, 
Semler shows how his radical ideas and strate¬ 
gies, as well as his nuts-and-bolts specifics, 
can be applied to any company anywhere. For ' 
instance, he includes actual forms like the 
questionnaires that Semco’s employees use to 
grade their managers. And he openly chal¬ 
lenges the way almost all corporations do 
business today. Destined to be read, re-read, 
and hotly discussed, MAVERICK is a landmark 
book about people and their companies, and 
how they both can achieve unimagined new 
heights of success. 

Ricardo Semler is the CEO of Semco S.A., a 
manufacturer of industrial equipment that 
employs several hundred people and that has 
moved from 56th to 4th in its industiy. Mr. Sem¬ 
ler is also a board member of SOS Atlantic . 
Forest, Brazil’s foremost environmental pro¬ 
tection entity, and regularly undertakes 
adventure travel: his trips include retracing 
Marco Polo’s route through Asia, ballooning 
over Tanzania, and rafting down the Nile. He 
lives in Sao Paulo, Brazil, with his wife. 

Jacket design by Bernadette Anthony 
Jacket photo by Paolo Fridman 

A A ’ 


A Time Warner Company 
< 1446-51696 1 



Semco’s last job posting generated 1,400 responses, adding to the 
company s pile of 2,000 unsolicited resumes of people waiting for a 
chance to work at this extraordinary company. In a recent poll, almost 
20 percent of all college students in Brazil said that they would like 
to work for Semco when they graduated. And at the same time its 
CEO, Ricardo Semler, was voted Business Leader of the Year by 
52,000 conservative businessmen, Brazil’s fearsome marxist Union 
leaders were calling him “the only trustworthy boss in the country.” 


is the true and fascinating story of Semco, a company that defies 
all conventional wisdom and yet is successful in one of the world’s 
most daunting economic climates. A story filled with wisdom and 
humor, told by those who led and lived through this business rev¬ 
olution, it is must reading for everyone in business—especially 
those looking for the workplace of their dreams.